JP5274663B2 - Photoelectrochemical cell and energy system using the same - Google Patents

Abstract

A photoelectrochemical cell (100) includes: a semiconductor electrode (120) including a substrate (121), a first n-type semiconductor layer (122) disposed on the substrate (121), and a second n-type semiconductor layer (123) and a conductor (124) disposed apart from each other on the first n-type semiconductor layer (122); a counter electrode (130) connected electrically to the conductor (124); an electrolyte (140) in contact with surfaces of the second n-type semiconductor layer (123) and the counter electrode (130); and a container (110) accommodating the semiconductor electrode (120), the counter electrode (130) and the electrolyte (140). In the semiconductor electrode (120), relative to a vacuum level, (I) band edge levels of a conduction band and a valence band in the second n-type semiconductor layer (123), respectively, are higher than band edge levels of a conduction band and a valence band in the first n-type semiconductor layer (122), (II) a Fermi level of the first n-type semiconductor layer (122) is higher than a Fermi level of the second n-type semiconductor layer (123), and (III) a Fermi level of the conductor (124) is higher than the Fermi level of the first n-type semiconductor layer (122). The photoelectrochemical cell (100) generates hydrogen by irradiation of the second n-type semiconductor layer (123) with light.

Description

  The present invention relates to a photoelectrochemical cell that decomposes water by light irradiation and an energy system using the same.

  Conventionally, by irradiating a semiconductor material functioning as a photocatalyst with light, water is decomposed to collect hydrogen and oxygen (see, for example, Patent Document 1 and Patent Document 2), or the surface of a substrate using the semiconductor material. It is known that the surface of the base material is made hydrophilic by coating the surface of the substrate (for example, see Patent Document 3).

Patent Document 1 discloses a technique for collecting hydrogen and oxygen from the surfaces of both electrodes by disposing an n-type semiconductor electrode and a counter electrode in an electrolytic solution and irradiating light on the surfaces of the n-type semiconductor electrode. Yes. Specifically, it is described that a TiO ₂ electrode, a ZnO electrode, a CdS electrode, or the like is used as the n-type semiconductor electrode.

  In Patent Document 2, a semiconductor electrode formed of a group III nitride semiconductor and a counter electrode are arranged in an electrolytic solution, and hydrogen and oxygen are generated from the surfaces of both electrodes by irradiating light on the surfaces of the semiconductor electrodes. A gas generating apparatus is disclosed.

  Patent Document 3 discloses a hydrophilic member comprising a base material and a film formed on the surface of the base material, wherein the film includes a titanium oxide layer containing titanium oxide particles, and the titanium oxide layer. And an island-shaped portion made of a second photocatalytic material other than titanium oxide, which is disposed above. Specifically, as the second photocatalytic material, the potential at the lower end of the conduction band and the upper end of the valence band is positive with respect to the standard hydrogen electrode potential as compared to titanium oxide (negative side with respect to the vacuum level). It is described that the material which exists in is used.

  In addition, as a photocatalytic thin film capable of obtaining highly efficient photocatalytic performance under natural light, at least one kind of ions of metal ions such as Nb, V, and Cr is implanted into the photocatalytic thin film produced on the substrate to obtain a band gap or a potential gradient. A photocatalytic thin film has also been proposed in which the film is changed in the thickness direction to form an inclined film (see Patent Document 4).

  In addition, a multilayer thin-film photocatalyst in which a first compound semiconductor layer and a second compound semiconductor layer having a band gap different from that of the first compound semiconductor layer are sequentially arranged on a conductive substrate is produced by hydrogen sulfide. There has also been proposed a technique for producing hydrogen by immersing the substrate in a solution containing hydrogen and irradiating the multilayer thin film photocatalyst with light (see Patent Document 5).

JP-A-51-123779 JP 2003-24764 A JP 2002-234105 A JP 2002-143688 A JP 2003-154272 A

  However, the techniques described in Patent Document 1 and Patent Document 2 have a problem that the quantum efficiency of the water decomposition reaction caused by light irradiation is low. This is because holes and electrons generated by photoexcitation have a high probability of recombining and disappearing before being used in the water electrolysis reaction.

  For example, in Patent Document 2, a group III nitride n-type semiconductor has a p-type semiconductor on the back surface, an electrode having the n-type semiconductor on the back surface, or an n-type semiconductor on the back surface of the p-type semiconductor, and a p-type semiconductor on the back surface. An electrode having a semiconductor is described, and the effect that charges are separated by an electromotive force generated by a pn junction is described. However, in such a configuration, carriers can be accumulated in the semiconductor on the outermost surface of the electrode, and a large charge separation effect cannot be expected.

  In Patent Document 3, among electrons and holes generated by photoexcitation, electrons move to the conduction band of the second photocatalytic material, and holes move to the valence band of titanium oxide, so that the electron-hole pair is separated. Therefore, it is described that the probability of recombination becomes low. However, Patent Document 3 does not describe anything about how the energy state at the joint surface between titanium oxide and the second photocatalytic material is set. When the junction surface between the titanium oxide and the second photocatalytic material is a Schottky junction, a Schottky barrier is generated in the conduction band and the valence band at the junction surface. At this time, among the electrons and holes generated by photoexcitation, the electrons are blocked by the Schottky barrier at the junction surface in the conduction band, and the Schottky barrier at the junction surface in the valence band functions as a hole reservoir. It accumulates near the junction surface of the electronic band. Therefore, there is a problem that the possibility of recombination of electrons and holes becomes higher than when titanium oxide and the second photocatalytic material are used alone.

  In Patent Document 4, the photocatalytic thin film is formed into an inclined film by metal ion doping. However, this configuration is a technique made for the purpose of improving the light utilization efficiency up to the visible light region by forming the photocatalytic thin film into an inclined film. For this reason, nothing is described as to how the energy state of the photocatalyst in the inclined film is set, and optimization such as charge separation is not performed.

  The multilayer thin film photocatalyst described in Patent Document 5 has a structure in which two semiconductors CdS and ZnS having different band gaps are joined, and this semiconductor ZnS and a conductive substrate Pt are joined. In Patent Document 5, since materials having different band gaps are bonded in this manner, electrons move along the gradient of the band gap to the semiconductor ZnS and further to the conductive base material Pt. It is described that it is easy to combine with hydrogen ions and easily generate hydrogen (paragraphs [0026] to [0027] in Patent Document 5). However, considering the Fermi level (vacuum reference value) of each material and paying attention to those junctions, the junction between CdS (−5.0 eV) and ZnS (−5.4 eV) is also ZnS ( Since the junction between −5.4 eV) and Pt (−5.7 eV) also has a lower Fermi level with respect to the direction of electron movement (the direction of movement from CdS to ZnS and further from ZnS to Pt), Schottky Barriers arise. Therefore, in this configuration, although electrons move along the band gap gradient, it is difficult to move smoothly.

  Therefore, in view of the above-described conventional problems, the present invention can efficiently separate electrons and holes generated by photoexcitation and can improve the quantum efficiency of a hydrogen generation reaction by light irradiation. It aims at providing a chemical cell and an energy system using the same.

In order to achieve the above object, the present invention provides:
A substrate, a first n-type semiconductor layer disposed on the substrate, and a second n-type semiconductor layer and a conductor disposed on the first n-type semiconductor layer so as to be spaced apart from each other; Including a semiconductor electrode;
A counter electrode electrically connected to the conductor;
An electrolyte in contact with the surface of the second n-type semiconductor layer and the counter electrode;
A container containing the semiconductor electrode, the counter electrode and the electrolyte;
With
In the semiconductor electrode, with reference to the vacuum level,
(I) The band edge levels of the conduction band and the valence band in the second n-type semiconductor layer are larger than the band edge levels of the conduction band and the valence band in the first n-type semiconductor layer, respectively. ,
(II) The Fermi level of the first n-type semiconductor layer is larger than the Fermi level of the second n-type semiconductor layer, and
(III) The Fermi level of the conductor is larger than the Fermi level of the first n-type semiconductor layer,
Provided is a first photoelectrochemical cell that generates hydrogen when the second n-type semiconductor layer is irradiated with light.

In order to achieve the above object, the present invention provides:
A substrate, a first p-type semiconductor layer disposed on the substrate, and a second p-type semiconductor layer and a conductor disposed on the first p-type semiconductor layer and spaced apart from each other; Including a semiconductor electrode;
A counter electrode electrically connected to the conductor;
An electrolyte in contact with the surface of the second p-type semiconductor layer and the counter electrode;
A container containing the semiconductor electrode, the counter electrode and the electrolyte;
With
In the semiconductor electrode, with reference to the vacuum level,
(I) The band edge levels of the conduction band and the valence band in the second p-type semiconductor layer are smaller than the band edge levels of the conduction band and the valence band in the first p-type semiconductor layer, respectively. ,
(II) the Fermi level of the first p-type semiconductor layer is smaller than the Fermi level of the second p-type semiconductor layer, and
(III) The Fermi level of the conductor is smaller than the Fermi level of the first p-type semiconductor layer,
A second photoelectrochemical cell that generates hydrogen by irradiating the second p-type semiconductor layer with light is provided.

  The energy system of the present invention is connected to the first or second photoelectrochemical cell of the present invention, the first or second photoelectrochemical cell, and a first pipe. A hydrogen storage device for storing hydrogen generated in the photoelectrochemical cell 2 and a fuel cell connected to the hydrogen storage device by a second pipe and converting the hydrogen stored in the hydrogen storage device into electric power And.

  According to the first and second photoelectrochemical cells of the present invention, since electrons and holes generated by photoexcitation can be efficiently separated by charge, the quantum efficiency of the hydrogen generation reaction by light irradiation can be improved. Since the energy system of the present invention includes such a photoelectrochemical cell, it can supply power efficiently.

Schematic which shows the structure of the photoelectrochemical cell of Embodiment 1 of this invention. In the photoelectrochemical cell of Embodiment 1 of this invention, the schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer In the photoelectrochemical cell of Embodiment 1 of this invention, the schematic diagram which shows the band structure after joining of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer The schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, the 1st n-type semiconductor layer, and the 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-1. The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-1 The schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, the 1st n-type semiconductor layer, and the 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-2. The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-2 The schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-3. The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-3 The schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-4. The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-4 The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-5 In the photoelectrochemical cell of the comparative form 1-5, the schematic diagram which shows the band structure after joining of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-6 In the photoelectrochemical cell of the comparative form 1-6, the schematic diagram which shows the band structure after joining of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-7 The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st n-type semiconductor layer, and a 2nd n-type semiconductor layer in the photoelectrochemical cell of the comparative form 1-7 Schematic which shows the structure of the photoelectrochemical cell of Embodiment 2 of this invention. In the photoelectrochemical cell of Embodiment 2 of this invention, the schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer Schematic diagram showing a band structure after bonding of a conductor constituting a semiconductor electrode, a first p-type semiconductor layer, and a second p-type semiconductor layer in the photoelectrochemical cell of Embodiment 2 of the present invention. The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-1. The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-1. The schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, the 1st p-type semiconductor layer, and the 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-2. The schematic diagram which shows the band structure after joining of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-2. The schematic diagram which shows the band structure before joining of the conductor which comprises a semiconductor electrode, the 1st p-type semiconductor layer, and the 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-3. The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-3 The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-4 The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-4 The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-5 In the photoelectrochemical cell of the comparative form 2-5, the schematic diagram which shows the band structure after joining of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-6 In the photoelectrochemical cell of the comparative form 2-6, the schematic which shows the band structure after joining of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer The schematic diagram which shows the band structure before the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-7 The schematic diagram which shows the band structure after the junction of the conductor which comprises a semiconductor electrode, a 1st p-type semiconductor layer, and a 2nd p-type semiconductor layer in the photoelectrochemical cell of the comparative form 2-7 Schematic which shows the structure of the photoelectrochemical cell of Embodiment 3 of this invention. Schematic which shows the structure of the photoelectrochemical cell of Embodiment 4 of this invention. Schematic which shows the structure of the energy system of Embodiment 5 of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The following embodiment is an example, and the present invention is not limited to the following embodiment. Moreover, in the following embodiment, the same code | symbol may be attached | subjected to the same member and the overlapping description may be abbreviate | omitted.

(Embodiment 1)
The structure of the photoelectrochemical cell of Embodiment 1 of this invention is demonstrated using FIGS. 1-3. FIG. 1 is a schematic diagram showing the configuration of the photoelectrochemical cell of the present embodiment. FIG. 2 is a schematic diagram showing a band structure before bonding of a conductor constituting a semiconductor electrode, a first n-type semiconductor layer, and a second n-type semiconductor layer in the photoelectrochemical cell of the present embodiment. . FIG. 3 is a schematic diagram showing a band structure after bonding of the conductor constituting the semiconductor electrode, the first n-type semiconductor layer, and the second n-type semiconductor layer in the photoelectrochemical cell of the present embodiment. . 2 and 3, the vertical axis indicates the energy level (unit: eV) based on the vacuum level.

  As shown in FIG. 1, the photoelectrochemical cell 100 of the present embodiment includes a semiconductor electrode 120, a counter electrode 130 that is a pair of electrodes with the semiconductor electrode 120, an electrolytic solution 140 containing water, a semiconductor electrode 120, And a container 110 having an opening that accommodates the counter electrode 130 and the electrolytic solution 140.

  Inside the container 110, the semiconductor electrode 120 and the counter electrode 130 are arranged so that the surfaces thereof are in contact with the electrolytic solution 140. The semiconductor electrode 120 includes a substrate 121, a first n-type semiconductor layer 122 disposed on the substrate 121, and a second n-type semiconductor disposed on the first n-type semiconductor layer 122 so as to be separated from each other. A layer 123 and a conductor 124. A portion of the container 110 that faces the second n-type semiconductor layer 123 of the semiconductor electrode 120 disposed in the container 110 (hereinafter abbreviated as a light incident portion 110a) is a material that transmits light such as sunlight. It consists of

  The conductor 124 and the counter electrode 130 in the semiconductor electrode 120 are electrically connected by a conducting wire 150. Here, the counter electrode means an electrode that exchanges electrons with the semiconductor electrode without using an electrolytic solution. Therefore, the counter electrode 130 in this embodiment is only required to be electrically connected to the conductor 124 constituting the semiconductor electrode 120, and the configuration such as the positional relationship with the semiconductor electrode 120 is not particularly limited. In this embodiment, since an n-type semiconductor is used for the semiconductor electrode 120, the counter electrode 130 serves as an electrode that receives electrons from the semiconductor electrode 120 without passing through the electrolytic solution 140.

  Next, the operation of the photoelectrochemical cell 100 according to the present embodiment will be described.

  When sunlight is irradiated from the light incident part 110a of the container 110 in the photoelectrochemical cell 100 to the second n-type semiconductor layer 123 of the semiconductor electrode 120 disposed in the container 110, the second n-type semiconductor layer is irradiated. In 123, electrons are generated in the conduction band and holes are generated in the valence band. The holes generated at this time move to the surface side of the second n-type semiconductor layer 123. Thereby, on the surface of the second n-type semiconductor layer 123, water is decomposed by the following reaction formula (1) to generate oxygen. On the other hand, electrons are bent at the band edge of the conduction band at the interface between the second n-type semiconductor layer 123 and the first n-type semiconductor layer 122 and at the interface between the first n-type semiconductor layer 122 and the conductor 124. And move to the conductor 124. The electrons that have moved to the conductor 124 move to the side of the counter electrode 130 that is electrically connected to the semiconductor electrode 120 via the conducting wire 150. Thereby, hydrogen is generated on the surface of the counter electrode 130 according to the following reaction formula (2).

4h ⁺ + 2H ₂ O → O ₂ ↑ + 4H ⁺ (1)
_{^{4e - + 4H + → 2H 2}} ↑ (2)

  Although details will be described later, since no Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123, the second n-type semiconductor layer is not hindered by electrons. 123 can move to the first n-type semiconductor layer 122. Further, since no Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 122 and the conductor 124, electrons can move from the first n-type semiconductor layer 122 to the conductor 124 without being blocked. Therefore, the probability of recombination of electrons and holes generated in the second n-type semiconductor layer 123 by photoexcitation is reduced. Thereby, according to the photoelectrochemical cell 100 of this Embodiment, the quantum efficiency of the hydrogen production | generation reaction by light irradiation can be improved.

  Next, the band structure of the conductor 124, the first n-type semiconductor layer 122, and the second n-type semiconductor layer 123 in the semiconductor electrode 120 will be described in detail. Note that the energy level of the band structure described here is based on the vacuum level. Hereinafter, the energy levels of the semiconductor and conductor band structures described in this specification are also based on the vacuum level.

As shown in FIG. 2, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 123 and the band edge level E _V2 of the valence band are respectively determined by the conduction of the first n-type semiconductor layer 122. It is larger than the band edge level E _C1 of the band and the band edge level E _V1 of the valence band.

Further, the Fermi level E _{F1 of} the first n-type semiconductor layer 122 is larger than the Fermi level E _{F2 of} the second n-type semiconductor layer 123, and the Fermi level E _Fc of the conductor 124 is the first level. It is larger than the Fermi level E _F1 of the n-type semiconductor layer 122.

Next, when the conductor 124, the first n-type semiconductor layer 122, and the second n-type semiconductor layer 123 are joined to each other, as shown in FIG. 3, the first n-type semiconductor layer 122 and the second n-type semiconductor layer 122 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 123, bending of the band edge occurs. At this time, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 123 and the band edge level E _V2 of the valence band respectively correspond to the band edges of the conduction band in the first n-type semiconductor layer 122. The Fermi level E _F1 of the first n-type semiconductor layer 123 is larger than the level E _C1 and the band edge level E _V1 of the valence band, and the Fermi level E _F1 of the second n-type semiconductor layer 123 is higher. _Since it is larger than _F 2, no Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123.

In addition, also at the junction surface between the conductor 124 and the first n-type semiconductor layer 122, the carrier moves so that the Fermi levels of each other coincide with each other, so that the band edge near the junction surface is bent. At this time, since the Fermi level E _Fc of the conductor 124 is larger than the Fermi level E _F1 of the first n-type semiconductor layer 122, the junction between the conductor 124 and the first n-type semiconductor layer 122 is Ohmic contact.

When the semiconductor electrode 120 as described above is brought into contact with the electrolytic solution 140, the band edge of the conduction band near the surface of the second n-type semiconductor layer 123 at the interface between the second n-type semiconductor layer 123 and the electrolytic solution 140. The level E _C2 and the band edge level E _V2 of the valence band are raised. Thereby, a space charge layer is generated near the surface of the second n-type semiconductor layer 123.

  As a comparative form, a form is assumed in which the band edge level of the conduction band in the second n-type semiconductor layer is smaller than the band edge level of the conduction band in the first n-type semiconductor layer. In this case, due to the bending of the band edge of the conduction band near the surface of the second n-type semiconductor layer and the difference in the band edge level of the conduction band between the first n-type semiconductor layer and the second n-type semiconductor layer. A well-type potential is generated at the band edge level of the conduction band inside the second n-type semiconductor layer. Due to this well-type potential, electrons accumulate in the second n-type semiconductor layer, and the probability that electrons and holes generated by photoexcitation recombine increases.

On the other hand, in the photoelectrochemical cell 100 of the present embodiment, the band edge level E _C2 of the conduction band in the second n-type semiconductor layer 123 is the band edge level of the conduction band in the first n-type semiconductor layer 122. Since it is set to be larger than the level E _C1 , the well-type potential as described above does not occur at the band edge level of the conduction band inside the second n-type semiconductor layer 123. Therefore, the electrons move to the first n-type semiconductor layer 122 side without accumulating inside the second n-type semiconductor layer 123, and the efficiency of charge separation is remarkably improved.

  Further, as another comparative form, it is assumed that the band edge level of the valence band in the second n-type semiconductor layer is smaller than the band edge level of the valence band in the first n-type semiconductor layer 122. To do. In this case, the difference between the band edge curve of the valence band near the surface of the second n-type semiconductor layer and the band edge level of the valence band between the first n-type semiconductor layer and the second n-type semiconductor layer. As a result, a well-type potential is generated at the band edge level of the valence band in the second n-type semiconductor layer. The holes generated inside the second n-type semiconductor layer by photoexcitation are divided and moved in the interface direction with the electrolytic solution and the interface direction with the first n-type semiconductor layer due to the well-type potential.

On the other hand, in the photoelectrochemical cell 100 of the present embodiment, the band edge level E _V2 of the valence band in the second n-type semiconductor layer 123 is equal to the valence band in the first n-type semiconductor layer 122. Since it is set to be larger than the band edge level E _V1 , the well-type potential as described above does not occur in the band edge level E _V2 of the valence band inside the second n-type semiconductor layer 123. . Therefore, the holes move in the direction of the interface with the electrolytic solution 140 without accumulating inside the second n-type semiconductor layer 123, so that the efficiency of charge separation is greatly improved.

Furthermore, in the photoelectrochemical cell 100 of the present embodiment, the Fermi level E _{F1 of} the first n-type semiconductor layer 122 is larger than the Fermi level E _{F2 of} the second n-type semiconductor layer 123. Is set. With this configuration, band bending occurs at the interface between the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123, and no Schottky barrier is generated. As a result, among the electrons and holes generated by photoexcitation inside the second n-type semiconductor layer 123, the electrons move to the conduction band of the first n-type semiconductor layer 122, and the holes change their valence band to the electrolyte 140. Therefore, electrons and holes are efficiently separated without being blocked by the Schottky barrier. As a result, the probability of recombination of electrons and holes generated inside the second n-type semiconductor layer 123 by photoexcitation is reduced, so that the quantum efficiency of the hydrogen generation reaction by light irradiation is improved.

  In the photoelectrochemical cell 100 of the present embodiment, the Fermi level of the conductor 124 is set to be higher than the Fermi level of the first n-type semiconductor layer 122. With this configuration, no Schottky barrier is generated even at the junction surface between the conductor 124 and the first n-type semiconductor layer 122. Therefore, the movement of electrons from the first n-type semiconductor layer 122 to the conductor 124 is not hindered by the Schottky barrier. As a result, the probability that electrons and holes generated inside the second n-type semiconductor layer 123 by photoexcitation are recombined is further reduced, and the quantum efficiency of the hydrogen generation reaction by light irradiation is further improved.

When the pH value of the electrolytic solution 140 is 0 and the temperature is 25 ° C., in this embodiment, the Fermi level E of the first n-type semiconductor layer 122 in the semiconductor electrode 120 in contact with the electrolytic solution 140 is used. _F1 is −4.44 eV or more, and the band edge level E _V2 of the valence band in the second n-type semiconductor layer 123 is −5.67 eV or less. When the semiconductor electrode 120 satisfies such an energy level, the Fermi level E _{Fc of the} conductor 124 in contact with the first n-type semiconductor layer 122 is −4.44 eV or more, which is a redox potential of hydrogen. It becomes. Thereby, hydrogen ions are efficiently reduced on the surface of the counter electrode 130 that is electrically connected to the conductor 124, so that hydrogen can be generated efficiently.

In addition, the band edge level E _V2 of the valence band in the second n-type semiconductor layer 123 becomes −5.67 eV or less which is the oxidation-reduction potential of water. Thereby, since water is efficiently oxidized on the surface of the second n-type semiconductor layer 123, oxygen can be generated efficiently.

As described above, the Fermi level E _F1 of the first n-type semiconductor layer 122 is set to −4.44 eV or more in the semiconductor electrode 120 in a state in which the pH value is 0 and the temperature is 25 ° C. In addition, when the band edge level E _V2 of the valence band in the second n-type semiconductor layer 123 is set to −5.67 eV or less, water can be efficiently decomposed.

In the present embodiment, the semiconductor electrode 120 satisfying the energy level as described above is shown. For example, the Fermi level E _{F1 of} the first n-type semiconductor layer 122 is less than −4.44 eV. Alternatively, the band edge level E _V2 of the valence band in the second n-type semiconductor layer 123 may exceed −5.67 eV. Even in such a case, hydrogen and oxygen can be generated.

  Here, the Fermi level and the potential at the bottom of the conduction band (band edge level) of the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123 are obtained using the flat band potential and the carrier concentration. it can. The flat band potential and carrier concentration of a semiconductor are obtained from a Mott-Schottky plot measured using the semiconductor to be measured as an electrode.

  In addition, the Fermi level of the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123 in a state of being in contact with the electrolyte solution 140 having a pH value of 0 and a temperature of 25 ° C. uses a semiconductor to be measured as an electrode. It can be obtained by measuring a Mott-Schottky plot in a state where the electrolyte solution having a pH value of 0 and a temperature of 25 ° C. is in contact with the semiconductor electrode.

  The potential (band edge level) at the top of the valence band of the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123 is the band gap and the first n-type semiconductor layer 122 obtained by the above method. And the potential at the lower end of the conduction band of the second n-type semiconductor layer 123. Here, the band gaps of the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123 are obtained from the light absorption edge observed in the light absorption spectrum measurement of the semiconductor to be measured.

  The Fermi level of the conductor 124 can be measured by, for example, photoelectron spectroscopy.

  Next, the material of each structural member provided in the photoelectrochemical cell 100 of the present embodiment will be described.

  The first n-type semiconductor layer 122 is preferably made of, for example, n-type gallium nitride (n-GaN). n-GaN has a pH value of 0 and can achieve a Fermi level of −4.44 eV or higher in a state where it is in contact with an electrolyte solution having a temperature of 25 ° C., and a particularly high quantum efficiency can be obtained for a water decomposition reaction. Furthermore, since n-GaN has a low sheet resistance, electrons can be efficiently extracted through the conductor 124 disposed thereon. Furthermore, according to n-GaN, high durability of the first n-type semiconductor layer 122 can also be realized.

  When the first n-type semiconductor layer 122 is formed of n-type gallium nitride (n-GaN), the second n-type semiconductor layer 123 includes at least one selected from the group consisting of gallium, indium, and aluminum. It is preferable to use an n-type group III nitride semiconductor containing one element as a group III element. That is, the second n-type semiconductor layer 123 includes an n-type gallium nitride / indium mixed crystal (n-GaInN), gallium nitride / aluminum mixed crystal (n-GaAlN), or gallium nitride / indium / aluminum mixed crystal (n -GaInAlN). These n-type Group III nitride semiconductors can realize a band edge level of −5.67 eV or less in the valence band in a state where the pH value is 0 and a temperature of 25 ° C. is in contact with the electrolyte, and the decomposition of water. A particularly high quantum efficiency is obtained for the reaction.

  When the first n-type semiconductor layer 122 is formed of n-GaN and the second n-type semiconductor layer 123 is formed of the n-type group III nitride semiconductor, the first n-type semiconductor layer 122 and the second n-type semiconductor layer 122 are formed. The n-type semiconductor layer 123 is preferably a crystal film obtained by epitaxial growth.

  According to the configuration of the semiconductor electrode in the photoelectrochemical cell of the present invention, even a semiconductor material that must be formed by epitaxial growth such as a GaN crystal film can be easily used by appropriately selecting the substrate 121. Therefore, the range of material selection is expanded.

  The first n-type semiconductor layer 122 may be formed of an n-type oxide semiconductor containing gallium, indium, and zinc. Such an oxide can achieve a Fermi level of −4.44 eV or higher in a state where it is in contact with an electrolyte solution having a pH value of 0 and a temperature of 25 ° C., so that a particularly high quantum efficiency can be obtained for a water decomposition reaction. . Furthermore, since such an oxide has a low sheet resistance, electrons can be efficiently taken out through the conductor 124 disposed thereon. Further, according to such an oxide, high durability of the first n-type semiconductor layer 122 can also be realized. Further, even if this oxide is amorphous, it can sufficiently exhibit the above performance, and for example, it can be formed by an inexpensive method such as sputtering or printing at room temperature.

  In the case where the first n-type semiconductor layer 122 is formed of an n-type oxide semiconductor containing gallium, indium, and zinc, the second n-type semiconductor layer 123 includes an oxide containing gallium, indium, and zinc. It is preferable to use an n-type semiconductor having a composition in which part of oxygen is substituted with nitrogen. An n-type semiconductor having such a composition can realize a band edge level of −5.67 eV or less in the valence band in a state where it is in contact with an electrolyte solution having a pH value of 0 and a temperature of 25 ° C. A particularly high quantum efficiency is obtained for the decomposition reaction. Further, an n-type semiconductor having such a composition can sufficiently exhibit the above performance even if it is amorphous, and can be formed by an inexpensive method such as sputtering or printing at room temperature, for example.

An n-type oxide semiconductor containing gallium, indium and zinc that can exhibit the above performance is, for example,
In _2x Ga _{2 (1-x)} O ₃ (ZnO) _y
(X and y satisfy 0.2 <x <1, 0.5 ≦ y, respectively)
It can be expressed as.

In the above formula, an oxide in which x is 0.5 is particularly preferable. By replacing part of the oxygen contained in the oxide to nitrogen, for example, a nitride containing gallium and indium _{(Ga z In 1-z N} (z is 0 satisfy <z <1)) can not be realized in the A high In content nitride can be realized. Further, this oxide has a wavelength corresponding to the band gap near 900 nm, and can absorb light having a wavelength up to about 900 nm. The band gap of the semiconductor used for water photolysis is 1.23 eV or more, and the wavelength corresponding to this is 1000 nm or less. Therefore, this oxide can be said to be optimal for the photolysis of water from the viewpoint of increasing the utilization efficiency of sunlight. Furthermore, in InGaO ₃ (ZnO) _y and InGaO ₃ (ZnO) _y , a nitride in which a part of oxygen is replaced with nitrogen is very stable. In addition to these, if ZnO is in solid solution, it becomes a relatively stable compound. In particular, when x> 0.2 is satisfied in the above formula, it is not possible to realize it with Ga _z In _1-z N. It becomes an optical semiconductor that can exhibit the effect of melting.

In the above formula, y preferably satisfies 1 ≦ y ≦ 6. By using such an oxide, a single-phase optical semiconductor can be easily obtained, so that excellent optical semiconductor performance can be realized. In particular, when y is 2 or 6 in the above formula, a part of the oxygen lattice can be completely replaced with nitrogen, so that a single phase optical semiconductor such as InGaZn ₂ N ₂ O ₂ or InGaZn ₆ N ₂ O ₆ is used. Can be obtained.

  In addition, a semiconductor material other than the above can be used for the first n-type semiconductor layer 122 and the second n-type semiconductor layer 123. For example, oxide, sulfide, selenide, telluride, nitride, oxynitride containing titanium, zirconium, vanadium, niobium, tantalum, chromium, molybdenum, tungsten, manganese, iron, cobalt, zinc or cadmium as a constituent element And phosphides.

  For example, as the first n-type semiconductor layer 122, an oxide containing titanium, zirconium, niobium, or zinc as a constituent element may be used. The first n-type semiconductor layer 122 may be a single substance of these oxides or a composite compound containing these oxides. In addition, an alkali metal ion, an alkaline earth metal, or the like may be added to these oxides.

  The carrier concentration of the second n-type semiconductor layer 123 is preferably lower than the carrier concentration of the first n-type semiconductor layer 122. The second n-type semiconductor layer 123 may be one selected from the group consisting of oxide, nitride, and oxynitride. In this manner, even when the second n-type semiconductor layer 123 is irradiated with light in a state where the semiconductor electrode 120 is in contact with the electrolytic solution 140, the second n-type semiconductor layer 123 is dissolved in the electrolytic solution 140. Therefore, the photoelectrochemical cell can be operated stably.

  For example, when titanium oxide is used as the first n-type semiconductor layer 122, for example, tantalum nitride, tantalum oxynitride, or cadmium sulfide can be used as the second n-type semiconductor layer 123, and among these, tantalum nitride or acid It is preferable to use tantalum nitride. In this manner, even when the second n-type semiconductor layer 123 is irradiated with light in a state where the semiconductor electrode 120 is in contact with the electrolytic solution 140, the second n-type semiconductor layer 123 is dissolved in the electrolytic solution. Therefore, the photoelectrochemical cell can be operated stably.

  The composition of the second n-type semiconductor layer 123 may be inclined along the thickness direction of the second n-type semiconductor layer 123. Note that, here, “inclined composition” means that the concentration of at least one element constituting the second n-type semiconductor layer 123 is along the thickness direction of the second n-type semiconductor layer 123. It means an increase or decrease. By tilting the composition of the second n-type semiconductor layer 123, electrons and holes move more smoothly in the second n-type semiconductor layer 123, and recombination between electrons and holes is less likely to occur. The effect is obtained. However, even when the composition of the second n-type semiconductor layer 123 is inclined, the relationship between the conduction edge and the band edge level of the valence band with the first n-type semiconductor layer 122, and the Fermi level. The relationship needs to satisfy the relationship defined in the photoelectrochemical cell of the present invention.

  In this embodiment mode, the conductor 124 of the semiconductor electrode 120 is in ohmic contact with the first n-type semiconductor layer 122. Therefore, for example, a metal such as Ti, Ni, Ta, Nb, Al, and Ag, or a conductive material such as ITO (Indium Tin Oxide) and FTO (Fluorine doped Tin Oxide) is used as the conductor 124. it can.

  A region of the surface of the conductor 124 that is not covered with the first n-type semiconductor layer 122 is preferably covered with an insulator such as a resin. According to such a configuration, the conductor 124 can be prevented from being dissolved in the electrolytic solution 140.

  An insulating substrate such as a sapphire substrate or a glass substrate is preferably used for the substrate 121 so that the movement of electrons from the first n-type semiconductor layer 122 to the conductor 124 is not affected. The first n-type semiconductor layer 122 is formed of n-GaN, the second n-type semiconductor layer 123 is formed of the n-type group III nitride semiconductor, and the first n-type semiconductor layer 122 and the second n-type semiconductor layer 122 are formed. When the n-type semiconductor layer 123 is a crystal film obtained by epitaxial growth, a sapphire substrate is used as the substrate 121.

  It is preferable to use a material with a small overvoltage for the counter electrode 130. In this embodiment, since an n-type semiconductor is used for the semiconductor electrode 120, hydrogen is generated at the counter electrode 130. Therefore, for example, Pt, Au, Ag, or Fe is preferably used as the counter electrode 130.

  The electrolyte solution 140 may be an electrolyte solution containing water. The electrolytic solution containing water may be acidic or alkaline. When a solid electrolyte is disposed between the semiconductor electrode 120 and the counter electrode 130, an electrolytic solution 140 that contacts the surface of the second n-type semiconductor layer 123 of the semiconductor electrode 120 and the surface of the counter electrode 130 is used as water for electrolysis. It is also possible to replace with pure water.

  Hereinafter, the photoelectrochemical cells of Comparative Examples 1-1 to 1-7 in which the energy levels of the first n-type semiconductor layer, the second n-type semiconductor layer, and the conductor are different from those of the semiconductor electrode 120 will be described. It shows and demonstrates the difference of the effect. In Comparative Examples 1-1 to 1-7 shown below, the relationship between the energy levels of the first n-type semiconductor layer, the second n-type semiconductor layer, and the conductor is the photoelectrochemical cell of this embodiment. Although the configuration is different from that of the photoelectrochemical cell 100, the configuration other than that is the same as that of the photoelectrochemical cell 100, and the description thereof is omitted.

<Comparison 1-1>
A photoelectrochemical cell according to Comparative Example 1-1 will be described with reference to FIGS. FIG. 4 is a schematic diagram showing a band structure before bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment. FIG. 5 is a schematic diagram showing a band structure after bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first n-type semiconductor layer 162 disposed on the substrate, a second n-type semiconductor layer 163 disposed on the first n-type semiconductor layer 162, and a conductive layer. And a body 164. As shown in FIG. 4, the semiconductor electrode of this comparative embodiment is such that the Fermi level E _{F1 of} the first n-type semiconductor layer 162 is smaller than the Fermi level E _{F2 of} the second n-type semiconductor layer 163. However, it is different from the semiconductor electrode 120 in the first embodiment.

Next, when the conductor 164, the first n-type semiconductor layer 162, and the second n-type semiconductor layer 163 are bonded to each other, as shown in FIG. 5, the first n-type semiconductor layer 162 and the second n-type semiconductor layer 162 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 163, bending of the band edge occurs. The band edge level E _C2 of the conduction band of the second n-type semiconductor layer 163 and the band edge level E _V2 of the valence band are respectively the band edge level E of the conduction band of the first n-type semiconductor layer 162. The Fermi level E _{F1 of} the first n-type semiconductor layer 162 is smaller than the Fermi level E _{F2 of} the second n-type semiconductor layer 163, which is larger than _C1 and the band edge level E _V1 of the valence band. Therefore, unlike the semiconductor electrode 120 of the first embodiment, a Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 162 and the second n-type semiconductor layer 163.

At the junction surface between the first n-type semiconductor layer 162 and the conductor 164, carriers move so that their Fermi levels coincide with each other, so that a band edge near the junction surface is bent. At this time, since the Fermi level E _Fc of the conductor 164 is larger than the Fermi level E _{F1 of} the first n-type semiconductor layer 162, the conductor 164 is similar to the semiconductor electrode 120 in the first embodiment. The junction with the first n-type semiconductor layer 162 is in ohmic contact.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 120 in Embodiment 1, a Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 162 and the second n-type semiconductor layer 163. This Schottky barrier prevents movement of electrons from the second n-type semiconductor layer 163 to the first n-type semiconductor layer 162. Therefore, in this comparative embodiment, the probability that electrons and holes generated in the second n-type semiconductor layer 163 by photoexcitation recombine is higher than that of the photoelectrochemical cell 100 according to Embodiment 1, and light irradiation is performed. The quantum efficiency of the hydrogen production reaction due to the decrease.

<Comparison 1-2>
A photoelectrochemical cell according to Comparative Example 1-2 will be described with reference to FIGS. FIG. 6 is a schematic diagram showing a band structure before bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment. FIG. 7 is a schematic diagram showing a band structure after bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first n-type semiconductor layer 262 disposed on the substrate, a second n-type semiconductor layer 263 disposed on the first n-type semiconductor layer 262, and a conductive layer. And a body 264. As shown in FIG. 6, the semiconductor electrode of this comparative embodiment has a Fermi level E _{F1 of} the first n-type semiconductor layer 262 smaller than the Fermi level E _{F2 of} the second n-type semiconductor layer 263, and The semiconductor device 120 is different from the semiconductor electrode 120 in Embodiment 1 in that the Fermi level E _Fc of the conductor 264 is smaller than the Fermi level E _{F1 of} the first n-type semiconductor layer 262.

When the conductor 264, the first n-type semiconductor layer 262, and the second n-type semiconductor layer 263 are joined to each other, as shown in FIG. 7, the first n-type semiconductor layer 262 and the second n-type semiconductor layer As the carriers move so that their Fermi levels coincide with each other at the joint surface with H.263, bending of the band edge occurs. At this time, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 263 and the band edge level E _V2 of the valence band are respectively the band edge of the conduction band of the first n-type semiconductor layer 262. Although the level E _C1 and the band edge level E _V1 of the valence band are larger, the Fermi level E _{F1 of} the first n-type semiconductor layer 262 is higher than the Fermi level E _{F2 of} the second n-type semiconductor layer 263. Therefore, unlike the semiconductor electrode 120 in Embodiment 1, a Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 262 and the second n-type semiconductor layer 263.

At the junction surface between the first n-type semiconductor layer 262 and the conductor 264, carriers move so that their Fermi levels coincide with each other, whereby a band edge in the vicinity of the junction surface of the first n-type semiconductor layer 262 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 264 is smaller than the Fermi level E _F1 of the first n-type semiconductor layer 262, the junction surface between the conductor 264 and the first n-type semiconductor layer 262 is used. A Schottky barrier occurs.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 120 in Embodiment 1, a Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 262 and the second n-type semiconductor layer 263. This Schottky barrier prevents movement of electrons from the second n-type semiconductor layer 263 to the first n-type semiconductor layer 262. Furthermore, in the case of the semiconductor electrode of this comparative embodiment, a Schottky barrier is also generated at the junction surface between the conductor 264 and the first n-type semiconductor layer 262. This Schottky barrier prevents movement of electrons from the first n-type semiconductor layer 262 to the conductor 264. Therefore, in the photoelectrochemical cell according to this comparative embodiment, the probability that electrons and holes generated in the second n-type semiconductor layer are recombined by photoexcitation is higher than that of the photoelectrochemical cell 100 of the first embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparison 1-3>
A photoelectrochemical cell according to Comparative Example 1-3 will be described with reference to FIGS. FIG. 8 is a schematic diagram showing a band structure before bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment. FIG. 9 is a schematic diagram showing a band structure after bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first n-type semiconductor layer 362 disposed on the substrate, a second n-type semiconductor layer 363 disposed on the first n-type semiconductor layer 362, and a conductive layer. And a body 364. As shown in FIG. 8, the semiconductor electrode of this comparative embodiment has a band edge level E _C2 of the conduction band of the second n-type semiconductor layer 363 such that the band edge level of the conduction band of the first n-type semiconductor layer 362 is the same. The difference from the semiconductor electrode 120 in the first embodiment is that it is smaller than the position E _C1 .

Next, when the conductor 364, the first n-type semiconductor layer 362, and the second n-type semiconductor layer 363 are bonded to each other, as shown in FIG. 9, the first n-type semiconductor layer 362 and the second n-type semiconductor layer 363 are joined. When the carriers move so that their Fermi levels coincide with each other at the interface with the semiconductor layer 363, the band edge is bent. At this time, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 363 is smaller than the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 362, and the first n-type Since the Fermi level E _F1 of the semiconductor layer 362 is larger than the Fermi level E _{F2 of} the second n-type semiconductor layer 363, the band edge of the conduction band is first from the second n-type semiconductor layer 363 side. Although it becomes smaller toward the junction surface with the n-type semiconductor layer 362, it increases from the junction surface toward the first n-type semiconductor layer 362 side.

At the junction surface between the first n-type semiconductor layer 362 and the conductor 364, the carrier moves so that the Fermi levels of the first n-type semiconductor layer 362 coincide with each other, whereby a band edge in the vicinity of the junction surface of the first n-type semiconductor layer 362 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 364 is larger than the Fermi level E _{F1 of} the first n-type semiconductor layer 362, the conductor 364 and the semiconductor electrode 120 in Embodiment 1 The junction with the first n-type semiconductor layer 362 is in ohmic contact.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 120 in Embodiment 1, the band edge of the conduction band is increased from the junction surface between the first n-type semiconductor layer 362 and the second n-type semiconductor layer 363. 1 increases toward the n-type semiconductor layer 362 side, movement of electrons from the second n-type semiconductor layer 363 to the first n-type semiconductor layer 362 is prevented. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second n-type semiconductor are recombined by photoexcitation is higher than that of the photoelectrochemical cell 100 of Embodiment 1, and the light The quantum efficiency of the hydrogen production reaction due to the irradiation of selenium decreases.

<Comparison 1-4>
A photoelectrochemical cell according to Comparative Embodiment 1-4 will be described with reference to FIGS. FIG. 10 is a schematic diagram showing a band structure before bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment. FIG. 11 is a schematic diagram showing a band structure after bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first n-type semiconductor layer 462 disposed on the substrate, a second n-type semiconductor layer 463 disposed on the first n-type semiconductor layer 462, and a conductive layer. And a body 464. As shown in FIG. 10, in the semiconductor electrode of this comparative embodiment, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 463 is equal to the band edge level of the conduction band of the first n-type semiconductor layer 462. The semiconductor electrode 120 in the first embodiment is that the Fermi level E _Fc of the conductor 464 is smaller than the level E _C1 and smaller than the Fermi level E _{F1 of} the first n-type semiconductor layer 462. Different.

Next, when the conductor 464, the first n-type semiconductor layer 462, and the second n-type semiconductor layer 463 are bonded to each other, as shown in FIG. 11, the first n-type semiconductor layer 462 and the second n-type semiconductor layer 462 are joined. As the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 463, bending of the band edge occurs. At this time, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 463 is smaller than the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 462, and the first n-type Since the Fermi level E _F1 of the semiconductor layer 462 is larger than the Fermi level E _{F2 of} the second n-type semiconductor layer 463, the band edge of the conduction band is the first from the second n-type semiconductor layer 463 side. Although it becomes smaller toward the junction surface with the n-type semiconductor layer 462, it increases from the junction surface toward the first n-type semiconductor layer 462 side.

At the junction surface between the first n-type semiconductor layer 462 and the conductor 464, carriers move so that their Fermi levels coincide with each other, whereby a band edge in the vicinity of the junction surface of the first n-type semiconductor layer 462 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 464 is smaller than the Fermi level E _F1 of the first n-type semiconductor layer 462, the junction between the conductor 464 and the first n-type semiconductor layer 462 is Schottky contact.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 120 in Embodiment 1, the band edge of the conduction band is first from the junction surface between the first n-type semiconductor layer 462 and the second n-type semiconductor layer 463. Since the size increases toward the first n-type semiconductor layer 462, the movement of electrons from the second n-type semiconductor layer 463 to the first n-type semiconductor layer 462 is prevented. Further, a Schottky barrier is generated at the junction surface between the conductor 464 and the first n-type semiconductor layer 462. This Schottky barrier prevents movement of electrons from the first n-type semiconductor layer 462 to the conductor 464. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second n-type semiconductor layer are recombined by photoexcitation is higher than that of the photoelectrochemical cell 100 of the first embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparative Form 1-5>
A photoelectrochemical cell according to Comparative Example 1-5 will be described with reference to FIGS. FIG. 12 is a schematic diagram showing a band structure before bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment. FIG. 13 is a schematic diagram showing a band structure after bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first n-type semiconductor layer 562 disposed on the substrate, a second n-type semiconductor layer 563 disposed on the first n-type semiconductor layer 562, and a conductive layer. And a body 564. As shown in FIG. 12, the semiconductor electrode of this comparative embodiment has a band edge level E _{C2 in} the conduction band of the second n-type semiconductor layer 563, and a band edge level in the conduction band of the first n-type semiconductor layer 562. The first embodiment is that the Fermi level E _{F1 of} the first n-type semiconductor layer 562 is smaller than the level E _C1 and smaller than the Fermi level E _{F2 of} the second n-type semiconductor layer 563. Unlike the semiconductor electrode 120 in FIG.

Next, when the conductor 564, the first n-type semiconductor layer 562, and the second n-type semiconductor layer 563 are bonded to each other, as illustrated in FIG. 13, the first n-type semiconductor layer 562 and the second n-type semiconductor layer 562 are connected to each other. At the interface with the n-type semiconductor layer 563, the carriers move so that their Fermi levels coincide with each other, so that the band edge is bent. At this time, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 563 is smaller than the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 562, and the first n-type Since the Fermi level E _F1 of the semiconductor layer 562 is smaller than the Fermi level E _{F2 of} the second n-type semiconductor layer 563, similarly to the semiconductor electrode 120 in the first embodiment, at the band edge of the conduction band, A Schottky barrier is not generated at the junction surface between the first n-type semiconductor layer 562 and the second n-type semiconductor layer 563. However, as shown in FIG. 13, the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 562 is larger than the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 563. Become.

At the junction surface between the first n-type semiconductor layer 562 and the conductor 564, the carrier moves so that the Fermi levels of the first n-type semiconductor layer 562 coincide with each other, whereby a band edge near the junction surface of the first n-type semiconductor layer 562 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 564 is larger than the Fermi level E _{F1 of} the first n-type semiconductor layer 562, the conductor 564 is similar to the semiconductor electrode 120 in Embodiment 1. The junction with the first n-type semiconductor layer 562 is in ohmic contact.

In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 120 in Embodiment 1, the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 562 is the conduction of the second n-type semiconductor layer 563. Since it becomes larger than the band edge level E _C2 of the band, the movement of electrons from the second n-type semiconductor layer 563 to the first n-type semiconductor layer 562 is prevented. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second n-type semiconductor layer are recombined by photoexcitation is higher than that of the photoelectrochemical cell 100 of the first embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparative Form 1-6>
A photoelectrochemical cell according to Comparative Embodiment 1-6 will be described with reference to FIGS. FIG. 14 is a schematic diagram showing a band structure before bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment. FIG. 15 is a schematic diagram showing a band structure after bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first n-type semiconductor layer 662 disposed on the substrate, a second n-type semiconductor layer 663 disposed on the first n-type semiconductor layer 662, and a conductive layer. And a body 664. As shown in FIG. 14, in the semiconductor electrode of this comparative embodiment, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 663 is equal to the band edge level of the conduction band of the first n-type semiconductor layer 662. The Fermi level E _{F1 of} the first n-type semiconductor layer 662 is smaller than the Fermi level E _{F2 of} the second n-type semiconductor layer 663 and the Fermi level of the conductor 664 is smaller than the level E _C1. The difference from the semiconductor electrode 120 in the first embodiment is that E _Fc is smaller than the Fermi level E _{F1 of} the first n-type semiconductor layer 662.

Next, when the conductor 664, the first n-type semiconductor layer 662, and the second n-type semiconductor layer 663 are bonded to each other, as shown in FIG. 15, the first n-type semiconductor layer 662 and the second n-type semiconductor layer 662 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 663, bending of the band edge occurs. At this time, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 663 is smaller than the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 662, and the first n-type Since the Fermi level E _F1 of the semiconductor layer 662 is smaller than the Fermi level E _{F2 of} the second n-type semiconductor layer 663, similarly to the semiconductor electrode 120 in Embodiment 1, at the band edge of the conduction band, A Schottky barrier is not generated at the junction surface between the first n-type semiconductor layer 662 and the second n-type semiconductor layer 663. However, as shown in FIG. 15, unlike the semiconductor electrode 120 in the first embodiment, the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 662 is the conduction of the second n-type semiconductor layer 663. It becomes larger than the band edge level E _C2 of the band.

At the junction surface between the first n-type semiconductor layer 662 and the conductor 664, carriers move so that their Fermi levels coincide with each other. As a result, the band edge near the bonding surface of the first n-type semiconductor layer 662 is bent. At this time, since the Fermi level E _Fc of the conductor 664 is smaller than the Fermi level E _F1 of the first n-type semiconductor layer 662, the junction surface between the conductor 664 and the first n-type semiconductor layer 662 is used. A Schottky barrier occurs.

In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 120 in Embodiment 1, the band edge level E _C1 of the conduction band of the first n-type semiconductor layer 662 is the conduction of the second n-type semiconductor layer 663. Since it becomes larger than the band edge level E _C2 of the band, the movement of electrons from the second n-type semiconductor layer 663 to the first n-type semiconductor layer 662 is prevented. Further, unlike the semiconductor electrode 120 in Embodiment 1, a Schottky barrier is generated at the junction surface between the conductor 664 and the first n-type semiconductor layer 662. This Schottky barrier prevents movement of electrons from the first n-type semiconductor layer 662 to the conductor 664. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second n-type semiconductor layer 663 are recombined by photoexcitation is higher than that of the photoelectrochemical cell 100 of the second embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparative Form 1-7>
A photoelectrochemical cell according to Comparative Example 1-7 will be described with reference to FIGS. FIG. 16 is a schematic diagram showing a band structure before bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment. FIG. 17 is a schematic diagram showing a band structure after bonding of the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first n-type semiconductor layer 762 disposed on the substrate, a second n-type semiconductor layer 763 disposed on the first n-type semiconductor layer 762, and a conductive layer. And a body 764. As shown in FIG. 16, the semiconductor electrode of this comparative embodiment is different from that of Embodiment 1 in that the Fermi level E _Fc of the conductor 764 is smaller than the Fermi level E _{F1 of} the first n-type semiconductor layer 762. Different from the semiconductor electrode 120.

When the conductor 764, the first n-type semiconductor layer 762, and the second n-type semiconductor layer 763 are joined to each other, as shown in FIG. 17, the first n-type semiconductor layer 762 and the second n-type semiconductor layer When the carriers move so that their Fermi levels coincide with each other at the joint surface with 763, bending of the band edge occurs. At this time, the band edge level E _C2 of the conduction band of the second n-type semiconductor layer 763 and the band edge level E _V2 of the valence band respectively correspond to the band edges of the conduction band in the first n-type semiconductor layer 762. The Fermi level E _F1 of the second n-type semiconductor layer 763 is greater than the level E _C1 and the band edge level E _V1 of the valence band, and the Fermi level E _{F1 of} the first n-type semiconductor layer 762 is greater. _Since it is larger than _F 2, no Schottky barrier is generated at the junction surface between the first n-type semiconductor layer 762 and the second n-type semiconductor layer 763.

At the junction surface between the first n-type semiconductor layer 762 and the conductor 764, carriers move so that their Fermi levels coincide with each other, whereby a band edge in the vicinity of the junction surface of the first n-type semiconductor layer 762 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 764 is smaller than the Fermi level E _F1 of the first n-type semiconductor layer 762, the junction surface between the conductor 764 and the first n-type semiconductor layer 762 is used. A Schottky barrier occurs.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 120 in Embodiment 1, a Schottky barrier is generated at the junction surface between the conductor 764 and the first n-type semiconductor layer 762. This Schottky barrier prevents movement of electrons from the first n-type semiconductor layer 762 to the conductor 764. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second n-type semiconductor layer 762 are recombined by photoexcitation is higher than that of the photoelectrochemical cell 100 of the first embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

(Embodiment 2)
The structure of the photoelectrochemical cell of Embodiment 2 of this invention is demonstrated using FIGS. 18-20. FIG. 18 is a schematic diagram showing the configuration of the photoelectrochemical cell of the present embodiment. FIG. 19 is a schematic diagram showing a band structure before bonding of a conductor constituting the semiconductor electrode, the first p-type semiconductor layer, and the second p-type semiconductor layer in the photoelectrochemical cell of the present embodiment. . FIG. 20 is a schematic diagram showing a band structure after bonding of the conductor constituting the semiconductor electrode, the first p-type semiconductor layer, and the second p-type semiconductor layer in the photoelectrochemical cell of the present embodiment. .

  As shown in FIG. 18, in the photoelectrochemical cell 200 according to the present embodiment, the configuration of the semiconductor electrode 220 is different from that of the semiconductor electrode 120 according to the first embodiment. The same as 100. Therefore, in the present embodiment, only the semiconductor electrode 220 will be described, and the same components as those in the photoelectrochemical cell 100 of the first embodiment will be denoted by the same reference numerals and description thereof will be omitted.

  The semiconductor electrode 220 is arranged so that the surface thereof is in contact with the electrolytic solution 140 as in the case of the first embodiment. The semiconductor electrode 220 includes a substrate 221, a first p-type semiconductor layer 222 disposed on the substrate 221, and a second p-type semiconductor disposed on the first p-type semiconductor layer 222 so as to be separated from each other. A layer 223 and a conductor 224. The second p-type semiconductor layer 223 faces the light incident part 110 a of the container 110.

  The conductor 224 in the semiconductor electrode 220 is electrically connected to the counter electrode 130 by a conducting wire 150.

  Next, the operation of the photoelectrochemical cell 200 of the present embodiment will be described.

  When sunlight is irradiated from the light incident part 110a of the container 110 in the photoelectrochemical cell 200 to the second p-type semiconductor layer 223 of the semiconductor electrode 220 disposed in the container 110, the second p-type semiconductor layer is irradiated. In 223, electrons are generated in the conduction band and holes are generated in the valence band. The holes generated at this time are valence band bands at the interface between the second p-type semiconductor layer 223 and the first p-type semiconductor layer 222 and at the interface between the first p-type semiconductor layer 222 and the conductor 224. It moves to the conductor 224 along the bending of the edge. The hole that has moved to the conductor 224 moves to the counter electrode 130 side that is electrically connected to the semiconductor electrode 220 via the conductive wire 150. Thereby, on the surface of the counter electrode 130, water is decomposed | disassembled by said reaction formula (1), and oxygen is generated. On the other hand, the electrons move to the surface side of the second p-type semiconductor layer 223 (interface side with the electrolytic solution 140). Thereby, hydrogen is generated by the reaction formula (2) on the surface of the second p-type semiconductor layer 223.

  Although details will be described later, since no Schottky barrier is generated at the junction surface between the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223, the second p-type semiconductor layer is not hindered. It can move from 223 to the first p-type semiconductor layer 222. In addition, since no Schottky barrier is generated at the junction surface between the conductor 224 and the first p-type semiconductor layer 222, holes can be moved from the first p-type semiconductor layer 222 to the conductor 224 without being blocked. Therefore, the probability of recombination of electrons and holes generated in the second p-type semiconductor layer 223 by photoexcitation is reduced. Thereby, according to the photoelectrochemical cell 200 of this Embodiment, the quantum efficiency of the hydrogen production | generation reaction by irradiation of light can be improved.

  Next, the band structure of the conductor 224, the first p-type semiconductor layer 222, and the second p-type semiconductor layer 223 in the semiconductor electrode 220 will be described in detail.

As shown in FIG. 19, the band edge level E _C2 of the conduction band of the second p-type semiconductor layer 223 and the band edge level E _V2 of the valence band are respectively determined by the conduction of the first p-type semiconductor layer 222. It is smaller than the band edge level E _C1 of the band and the band edge level E _V1 of the valence band.

The Fermi level E _F1 of the first p-type semiconductor layer 222 is smaller than the Fermi level E _{F2 of} the second p-type semiconductor layer 223, and the Fermi level E _Fc of the conductor 224 is It is smaller than the Fermi level E _F1 of the p-type semiconductor layer 222.

Next, when the conductor 224, the first p-type semiconductor layer 222, and the second p-type semiconductor layer 223 are joined to each other, as shown in FIG. 20, the first p-type semiconductor layer 222 and the second p-type semiconductor layer 222 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 223, bending of the band edge occurs. The band edge level E _C2 of the conduction band of the second p-type semiconductor layer 223 and the band edge level E _V2 of the valence band are respectively the band edge level E of the conduction band of the first p-type semiconductor layer 222. _C1 and the band edge level E _V1 of the valence band, and the Fermi level E _{F1 of} the first p-type semiconductor layer 222 is smaller than the Fermi level E _{F2 of} the second p-type semiconductor layer 223. Therefore, no Schottky barrier is generated at the junction surface between the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223.

Further, carriers move so that the Fermi levels of the first p-type semiconductor layer 222 and the conductor 224 coincide with each other, so that the first p-type semiconductor layer 222 near the joint surface. The band edge is bent. Since the Fermi level E _Fc of the conductor 224 is smaller than the Fermi level E _{F1 of} the first p-type semiconductor layer 222, the junction between the conductor 224 and the first p-type semiconductor layer 222 is an ohmic contact. Become.

When the semiconductor electrode 220 as described above is brought into contact with the electrolytic solution 140, the band edge of the conduction band near the surface of the second p-type semiconductor layer 223 at the interface between the second p-type semiconductor layer 223 and the electrolytic solution 140. The level E _C2 and the band edge level E _V2 of the valence band are lowered. Thereby, a space charge layer is generated near the surface of the second p-type semiconductor layer 223.

  As a comparative form, a form is assumed in which the band edge level of the conduction band in the second p-type semiconductor layer is larger than the band edge level of the conduction band in the first p-type semiconductor layer. In this case, due to the bending of the band edge of the conduction band near the surface of the second p-type semiconductor layer and the difference in the band edge level of the conduction band between the first p-type semiconductor layer and the second p-type semiconductor layer. A well-type potential is generated at the band edge level of the conduction band inside the second p-type semiconductor layer. Electrons generated inside the second p-type semiconductor layer by photoexcitation are divided and moved in the interface direction with the electrolytic solution and the interface direction with the first p-type semiconductor layer due to the well-type potential.

On the other hand, in the photoelectrochemical cell 200 of the present embodiment, the band edge level E _C2 of the conduction band in the second p-type semiconductor layer 223 is the band edge level of the conduction band in the first p-type semiconductor layer 222. Since it is set to be smaller than the level E _C1 , the well-type potential as described above does not occur at the band edge level of the conduction band inside the second p-type semiconductor layer 223. For this reason, the electrons in the second p-type semiconductor layer 223 move in the direction of the interface with the electrolytic solution 140, so that the efficiency of charge separation is significantly improved.

  As another comparative mode, a mode is assumed in which the band edge level of the valence band in the second p-type semiconductor layer is larger than the band edge level of the valence band in the first p-type semiconductor layer. In this case, the difference between the band edge curve of the valence band near the surface of the second p-type semiconductor layer and the band edge level of the valence band between the first p-type semiconductor layer and the second p-type semiconductor layer. As a result, a well-type potential is generated at the band edge level of the valence band inside the second p-type semiconductor layer. Due to the well-type potential, holes generated in the second p-type semiconductor layer by photoexcitation accumulate in the second p-type semiconductor layer.

On the other hand, in the photoelectrochemical cell 200 of the present embodiment, the band edge level E _V2 of the valence band in the second p-type semiconductor layer 223 is equal to the valence band in the first p-type semiconductor layer 222. Since it is set to be smaller than the band edge level E _V1 , the well-type potential as described above does not occur in the band edge level of the valence band in the second p-type semiconductor layer 223. Therefore, the holes move in the direction of the interface with the first p-type semiconductor layer 222 without accumulating inside the second p-type semiconductor layer 223, so that the charge separation efficiency is remarkably improved.

In the photoelectrochemical cell 200 of the present embodiment, the band edge levels of the conduction band and the valence band of the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223 have the band edge levels. In addition to the above setting, the Fermi level E _{F1 of} the first p-type semiconductor layer 222 is set to be smaller than the Fermi level E _{F2 of} the second p-type semiconductor layer 223. ing. With this configuration, band bending occurs at the interface between the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223, and no Schottky barrier is generated. As a result, among the electrons and holes generated by photoexcitation inside the second p-type semiconductor layer 223, the electrons move in the conduction band toward the interface with the electrolytic solution 140, and the holes are formed in the first p-type semiconductor layer 222. Move to the valence band. That is, electrons and holes are efficiently separated by electric charges without being blocked by the Schottky barrier. As a result, the probability of recombination of electrons and holes generated inside the second p-type semiconductor layer 223 by photoexcitation is reduced, so that the quantum efficiency of the hydrogen generation reaction by light irradiation is improved.

Further, in the photoelectrochemical cell 200 of the present embodiment, the Fermi level E _Fc of the conductor 224 is set to be smaller than the Fermi level E _{F1 of} the first p-type semiconductor layer 222. . With this configuration, no Schottky barrier is generated even at the junction surface between the conductor 224 and the first p-type semiconductor layer 222. Therefore, the movement of holes from the first p-type semiconductor layer 222 to the conductor 224 is not hindered by the Schottky barrier, so that electrons and holes generated inside the second p-type semiconductor layer 223 by photoexcitation are generated. The probability of recombination is further reduced, and the quantum efficiency of the hydrogen generation reaction by light irradiation is further improved.

Further, when the pH value of the electrolytic solution 140 is 0 and the temperature is 25 ° C., in this embodiment, the Fermi level of the first p-type semiconductor layer 222 in the semiconductor electrode 220 in contact with the electrolytic solution 140 is used. The level E _F1 is −5.67 eV or less, and the band edge level E _C2 of the conduction band in the second p-type semiconductor layer 223 is −4.44 eV or more. When the semiconductor electrode 220 satisfies such an energy level, the Fermi level E _{Fc of the} conductor 224 that is in contact with the first p-type semiconductor layer 222 is −5.67 eV, which is a redox potential of water. It becomes as follows. Therefore, since water is efficiently oxidized on the surface of the counter electrode 130 that is electrically connected to the conductor 224, oxygen can be generated efficiently.

In addition, the band edge level E _C2 of the conduction band in the second p-type semiconductor layer 223 is −4.44 eV or more which is the oxidation-reduction potential of hydrogen. Accordingly, hydrogen ions are efficiently reduced on the surface of the second p-type semiconductor layer 223, so that hydrogen can be generated efficiently.

As described above, the Fermi level E _F1 of the first p-type semiconductor layer 222 is set to −5.67 eV or less in the semiconductor electrode 220 in a state where it is in contact with the electrolytic solution 140 having a pH value of 0 and a temperature of 25 ° C. In addition, when the band edge level E _C2 of the conduction band in the second p-type semiconductor layer 223 is set to −4.44 eV or more, water can be efficiently decomposed.

In the present embodiment, the semiconductor electrode 220 satisfying the energy level as described above is shown. However, for example, the Fermi level E _{F1 of} the first p-type semiconductor layer 222 exceeds −5.67 eV. Alternatively, the band edge level E _C2 of the conduction band in the second p-type semiconductor layer 223 may be less than −4.44 eV. Even in such a case, hydrogen and oxygen can be generated.

  Here, the Fermi level and the potential at the top of the valence band (band edge level) of the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223 are obtained using the flat band potential and the carrier concentration. Can do. The flat band potential and carrier concentration of a semiconductor are obtained from a Mott-Schottky plot measured using the semiconductor to be measured as an electrode.

  The Fermi level of the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223 in contact with the electrolytic solution 140 having a pH value of 0 and a temperature of 25 ° C. uses the semiconductor to be measured as an electrode. It can be obtained by measuring a Mott-Schottky plot in a state where the electrolyte solution having a pH value of 0 and a temperature of 25 ° C. is in contact with the semiconductor electrode.

  The potential (band edge level) at the lower end of the conduction band of the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223 is the band gap, the first p-type semiconductor layer 222 obtained by the above method, and It can be obtained using the potential (band edge level) at the top of the valence band of the second p-type semiconductor layer 223. Here, the band gaps of the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223 are obtained from the light absorption edge observed in the light absorption spectrum measurement of the semiconductor to be measured.

  The Fermi level of the conductor 224 can be measured by a method similar to that in Embodiment 1.

  Next, materials of the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223 in this embodiment will be described.

  The first p-type semiconductor layer 222 is preferably made of p-type gallium nitride (p-GaN). p-GaN can achieve a Fermi level of −5.67 eV or less in a state where it is in contact with an electrolyte solution having a pH value of 0 and a temperature of 25 ° C., and a particularly high quantum efficiency can be obtained with respect to the water decomposition reaction. Furthermore, since p-GaN has a low sheet resistance, electrons are efficiently injected through the conductor 224 disposed thereon. Furthermore, according to p-GaN, high durability of the first p-type semiconductor layer 222 can also be realized.

  The second p-type semiconductor layer 223 is preferably made of a p-type group III nitride semiconductor containing gallium and at least one element selected from the group consisting of indium and aluminum as a group III element. That is, the second p-type semiconductor layer 223 includes a p-type gallium nitride / indium mixed crystal (n-GaInN), gallium nitride / aluminum mixed crystal (n-GaAlN), or gallium nitride / indium / aluminum mixed crystal (n -GaInAlN). These p-type Group III nitride semiconductors can realize a band edge level of −4.44 eV or higher in the conduction band in a state where the pH value is 0 and a temperature of 25 ° C. is in contact with the electrolyte, and the water decomposition reaction A particularly high quantum efficiency for is obtained.

  When the first p-type semiconductor layer 222 is formed of p-GaN and the second p-type semiconductor layer 223 is formed of the p-type group III nitride semiconductor, the first p-type semiconductor layer 222 and the second p-type semiconductor layer 222 are formed. The p-type semiconductor layer 223 is desirably a crystal film obtained by epitaxial growth.

  A semiconductor material other than the above can be used for the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223. For example, the first p-type semiconductor layer 222 and the second p-type semiconductor layer 223 include oxides, sulfides, selenides, tellurium containing copper, silver, gallium, indium, germanium, tin, antimony, or the like as constituent elements. A nitride, nitride, oxynitride, phosphide, or the like can be used.

Among these, as the first p-type semiconductor layer 222, it is preferable to use a copper oxide. In this way, the Fermi level E _F2 of the first p-type semiconductor layer 222 can be set to −5.67 eV or less in a state where it is in contact with the electrolytic solution having a pH value of 0 and a temperature of 25 ° C. The first p-type semiconductor layer 222 may be a simple substance of copper oxide or a complex compound containing copper oxide. Moreover, what added metal ions other than copper to the above compound may be used.

  The carrier concentration of the second p-type semiconductor layer 223 is preferably lower than the carrier concentration of the first p-type semiconductor layer 222. The second p-type semiconductor layer 223 is preferably one selected from the group consisting of oxide, nitride, and oxynitride. In this manner, even when the second p-type semiconductor layer 223 of the semiconductor electrode 220 is irradiated with light in a state where the semiconductor electrode 220 is in contact with the electrolytic solution 140, the second p-type semiconductor layer 223 is not dissolved in the electrolytic solution. No dissolution in 140. Therefore, the photoelectrochemical cell can be operated stably.

  When copper oxide is used for the first p-type semiconductor layer 222, for example, copper indium sulfide can be used for the second p-type semiconductor layer 223.

  The composition of the second p-type semiconductor layer 223 may be inclined along the thickness direction of the second p-type semiconductor layer 223. Here, the composition is inclined that the concentration of at least one element constituting the second p-type semiconductor layer 223 is along the thickness direction of the second p-type semiconductor layer 223. It means an increase or decrease. By tilting the composition of the second p-type semiconductor layer 223, electrons and holes move more smoothly in the second p-type semiconductor layer 223, and recombination of electrons and holes is less likely to occur. The effect is obtained. However, even when the composition of the second p-type semiconductor layer 223 is tilted, the relationship between the conduction edge and the band edge level of the valence band with the first p-type semiconductor layer, and the relationship between the Fermi level Needs to satisfy the relationship defined in the photoelectrochemical cell of the present invention.

  For the conductor 224, for example, a metal such as Ti, Ni, Ta, Nb, Al, and Ag, or a conductive material such as ITO and FTO can be used. From these, a material that makes ohmic contact with the first p-type semiconductor layer 222 may be appropriately selected.

  A region of the surface of the conductor 224 that is not covered with the first p-type semiconductor layer 222 is preferably covered with an insulator such as a resin. According to such a configuration, the conductor 224 can be prevented from being dissolved in the electrolytic solution 140.

  An insulating substrate such as a sapphire substrate or a glass substrate is preferably used for the substrate 221 so as not to affect the movement of holes from the first p-type semiconductor layer 222 to the conductor 224. The first p-type semiconductor layer 222 is formed of p-GaN, the second p-type semiconductor layer 223 is formed of the p-type group III nitride semiconductor, and the first p-type semiconductor layer 222 and the second p-type semiconductor layer 222 are formed. When the p-type semiconductor layer 223 is a crystal film obtained by epitaxial growth, a sapphire substrate is used as the substrate 221.

  Hereinafter, the photoelectrochemical cells of Comparative Examples 2-1 to 2-7 in which the energy levels of the first p-type semiconductor layer, the second p-type semiconductor layer, and the conductor are different from those of the semiconductor electrode 220 will be described. It shows and demonstrates the difference of the effect. In Comparative Examples 2-1 to 2-7 shown below, the relationship between the energy levels of the first p-type semiconductor layer, the second p-type semiconductor layer, and the conductor is the photoelectrochemical cell of this embodiment. Although the configuration is different from that of the photoelectrochemical cell 200, the other configuration is the same as that of the photoelectrochemical cell 200, and the description thereof is omitted.

<Comparison 2-1>
A photoelectrochemical cell according to Comparative Example 2-1 will be described with reference to FIGS. FIG. 21 is a schematic diagram showing a band structure before bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment. FIG. 22 is a schematic diagram showing a band structure after bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first p-type semiconductor layer 172 disposed on the substrate, a second p-type semiconductor layer 173 disposed on the first p-type semiconductor layer 172, and a conductive layer. And a body 174. As shown in FIG. 21, the semiconductor electrode of this comparative embodiment is such that the Fermi level E _{F1 of} the first p-type semiconductor layer 172 is larger than the Fermi level E _{F2 of} the second p-type semiconductor layer 173. However, it is different from the semiconductor electrode 220 in the second embodiment.

Next, when the conductor 174, the first p-type semiconductor layer 172, and the second p-type semiconductor layer 173 are joined together, as shown in FIG. 22, the first p-type semiconductor layer 172 and the second p-type semiconductor layer 172 are connected. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the semiconductor layer 173, the band edge is bent. At this time, the band edge level E _C2 of the conduction band of the second p-type semiconductor layer 173 and the band edge level E _V2 of the valence band are respectively the band edges of the conduction band of the first p-type semiconductor layer 172. is smaller than the band edge level E _V1 of level E _C1 and the valence band, the Fermi level E _F1 of the first p-type semiconductor layer 172 is higher than the Fermi level E _F2 of the second p-type semiconductor layer 173 Therefore, unlike the semiconductor electrode 220 in the second embodiment, a Schottky barrier is generated at the junction surface between the first p-type semiconductor layer 172 and the second p-type semiconductor layer 173.

At the junction surface between the first p-type semiconductor layer 172 and the conductor 174, the carrier moves so that the Fermi levels of the first p-type semiconductor layer 172 coincide with each other, so that the band edge near the junction surface of the first p-type semiconductor layer 172 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 174 is smaller than the Fermi level E _{F1 of} the first p-type semiconductor layer 172, similarly to the semiconductor electrode 220 in Embodiment 2, the conductor 174 and the second The junction with the first p-type semiconductor layer 172 is in ohmic contact.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 220 in Embodiment 2, a Schottky barrier is generated at the junction surface between the first p-type semiconductor layer 172 and the second p-type semiconductor layer 173. This Schottky barrier prevents the movement of holes from the second p-type semiconductor layer 173 to the first p-type semiconductor layer 172. Therefore, in the photoelectrochemical cell according to this comparative embodiment, the probability that electrons and holes generated in the second p-type semiconductor layer 173 are recombined by photoexcitation is higher than that in the photoelectrochemical cell 200 of the second embodiment. Thus, the quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparison 2-2>
A photoelectrochemical cell according to Comparative Example 2-2 will be described with reference to FIGS. FIG. 23 is a schematic diagram showing a band structure before bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment. FIG. 24 is a schematic diagram showing a band structure after bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment is composed of a conductor 274, a first p-type semiconductor layer 272, and a second p-type semiconductor layer 273. As shown in FIG. 23, the semiconductor electrode of this comparative embodiment has a Fermi level E _{F1 of} the first p-type semiconductor layer 272 larger than the Fermi level E _{F2 of} the second p-type semiconductor layer 273, and It differs from the semiconductor electrode 220 in Embodiment 2 in that the Fermi level E _Fc of the conductor 274 is larger than the Fermi level E _{F1 of} the first p-type semiconductor layer 272.

Next, when the conductor 274, the first p-type semiconductor layer 272, and the second p-type semiconductor layer 273 are joined to each other, as shown in FIG. 24, the first p-type semiconductor layer 272 and the second p-type semiconductor layer 272 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 273, bending of the band edge occurs. At this time, the band edge level E _C2 of the conduction band of the second p-type semiconductor layer 273 and the band edge level E _V2 of the valence band are respectively the band edges of the conduction band of the first p-type semiconductor layer 272. is smaller than the band edge level E _V1 of level E _C1 and the valence band, the Fermi level E _F1 of the first p-type semiconductor layer 272 than the Fermi level E _F2 of the second p-type semiconductor layer 273 Therefore, unlike the semiconductor electrode 220 in the second embodiment, a Schottky barrier is generated at the junction surface between the first p-type semiconductor layer 272 and the second p-type semiconductor layer 273.

At the junction surface between the first p-type semiconductor layer 272 and the conductor 274, carriers move so that their Fermi levels coincide with each other, whereby a band edge in the vicinity of the junction surface of the first p-type semiconductor layer 272 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 274 is larger than the Fermi level E _F1 of the first p-type semiconductor layer 272, the junction between the conductor 274 and the first p-type semiconductor layer 272 is Schottky contact.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 220 in Embodiment 2, a Schottky barrier is generated at the junction surface between the first p-type semiconductor layer 272 and the second p-type semiconductor layer 273. This Schottky barrier prevents the movement of holes from the second p-type semiconductor layer 273 to the first p-type semiconductor layer 272. Further, in the case of this comparative embodiment, a Schottky barrier is also generated at the junction surface between the conductor 274 and the first p-type semiconductor layer 272. This Schottky barrier prevents the movement of holes from the first p-type semiconductor layer 272 to the conductor 274. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second p-type semiconductor layer 273 are recombined by photoexcitation is higher than that of the photoelectrochemical cell 200 of the second embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparison 2-3>
A photoelectrochemical cell according to Comparative Example 2-3 will be described with reference to FIGS. FIG. 25 is a schematic diagram showing a band structure before bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment. FIG. 26 is a schematic diagram showing a band structure after bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first p-type semiconductor layer 372 disposed on the substrate, a second p-type semiconductor layer 373 disposed on the first p-type semiconductor layer 372, and a conductive layer. And a body 374. As shown in FIG. 25, the semiconductor electrode of this comparative embodiment has a band edge level E _{V2 in} the valence band of the second p-type semiconductor layer 373, and a band in the valence band of the first p-type semiconductor layer 372. It differs from the semiconductor electrode 220 in the second embodiment in that it is larger than the edge level E _V1 .

Next, when the conductor 374, the first p-type semiconductor layer 372, and the second p-type semiconductor layer 373 are joined together, as shown in FIG. 26, the first p-type semiconductor layer 372 and the second p-type semiconductor layer 372 are joined. As the carriers move so that their Fermi levels coincide with each other at the joint surface with the semiconductor layer 373, the band edge is bent. At this time, the band edge level E _V2 of the valence band of the second p-type semiconductor layer 373 is larger than the band edge level E _V1 of the valence band of the first p-type semiconductor layer 372, and the first Since the Fermi level E _F1 of the p-type semiconductor layer 372 is smaller than the Fermi level E _{F2 of} the second p-type semiconductor layer 373, the band edge of the valence band is on the second p-type semiconductor layer 373 side. Increases from the junction surface toward the first p-type semiconductor layer 372, but decreases from the junction surface toward the first p-type semiconductor layer 372.

At the junction surface between the first p-type semiconductor layer 372 and the conductor 374, the carrier moves so that the Fermi levels of the first p-type semiconductor layer 372 coincide with each other, whereby a band edge in the vicinity of the junction surface of the first p-type semiconductor layer 372 is obtained. Bending occurs. Since the Fermi level E _Fc of the conductor 374 is smaller than the Fermi level E _{F1 of} the first p-type semiconductor layer 372, similarly to the semiconductor electrode 220 in Embodiment 2, the conductor 374 and the first The junction with the p-type semiconductor layer 372 is in ohmic contact.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 220 in Embodiment 2, the band edge of the valence band is increased from the junction surface between the first p-type semiconductor layer 372 and the second p-type semiconductor layer 373. Since the first p-type semiconductor layer 372 becomes smaller, the movement of holes from the second p-type semiconductor layer 373 to the first p-type semiconductor layer 372 is prevented. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second p-type semiconductor layer 373 are recombined by photoexcitation is higher than that of the photoelectrochemical cell 200 of the second embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparison 2-4>
A photoelectrochemical cell according to Comparative Example 2-4 will be described with reference to FIGS. FIG. 27 is a schematic diagram showing a band structure before bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment. FIG. 28 is a schematic diagram showing a band structure after bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first p-type semiconductor layer 472 disposed on the substrate, a second p-type semiconductor layer 473 disposed on the first p-type semiconductor layer 472, and a conductive layer. And a body 474. As shown in FIG. 27, the semiconductor electrode of this comparative embodiment has a band edge level E _{V2 in} the valence band of the second p-type semiconductor layer 473, and a band in the valence band of the first p-type semiconductor layer 472. The semiconductor electrode 220 according to the second embodiment is larger than the edge level E _V1 and has a Fermi level E _Fc of the conductor 474 larger than the Fermi level E _{F1 of} the first p-type semiconductor layer 472. And different.

Next, when the conductor 474, the first p-type semiconductor layer 472, and the second p-type semiconductor layer 473 are bonded to each other, as shown in FIG. 28, the first p-type semiconductor layer 472 and the second p-type semiconductor layer 472 are joined. As the carriers move so that their Fermi levels coincide with each other at the junction surface of the type semiconductor layer 473, the band edge is bent. The band edge level E _V2 of the valence band of the second p-type semiconductor layer 473 is larger than the band edge level E _V1 of the valence band of the first p-type semiconductor layer 472, and the first p-type semiconductor Since the Fermi level E _F1 of the layer 472 is smaller than the Fermi level E _{F2 of} the second p-type semiconductor layer 473, the band edge of the valence band is first from the second p-type semiconductor layer 473 side. Although it becomes larger toward the junction surface with the p-type semiconductor layer 472, it decreases from the junction surface toward the first p-type semiconductor layer 472 side.

At the junction surface between the first p-type semiconductor layer 472 and the conductor 474, carriers move so that their Fermi levels coincide with each other, whereby a band edge in the vicinity of the junction surface of the first p-type semiconductor layer 472 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 474 is larger than the Fermi level E _F1 of the first p-type semiconductor layer 472, the junction between the conductor 474 and the first p-type semiconductor layer 472 is Schottky contact.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 220 in Embodiment 2, the band edge of the valence band is increased from the junction surface between the first p-type semiconductor layer 472 and the second p-type semiconductor layer 473. Since the first p-type semiconductor layer 472 becomes smaller toward the p-type semiconductor layer 472, the movement of holes from the second p-type semiconductor layer 473 to the first p-type semiconductor layer 472 is prevented. Further, in the case of this comparative embodiment, a Schottky barrier is generated at the junction surface between the conductor 474 and the first p-type semiconductor layer 472. This Schottky barrier prevents the movement of holes from the first p-type semiconductor layer 472 to the conductor 474. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second p-type semiconductor layer are recombined by photoexcitation is higher than that of the photoelectrochemical cell 400 of the second embodiment, The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

<Comparative Form 2-5>
A photoelectrochemical cell according to Comparative Example 2-5 will be described with reference to FIGS. FIG. 29 is a schematic diagram showing a band structure before bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment. FIG. 30 is a schematic diagram showing a band structure after bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a conductor 574, a first p-type semiconductor layer 572, and a second p-type semiconductor layer 573. As shown in FIG. 29, in the semiconductor electrode of this comparative embodiment, the band edge level E _V2 of the valence band of the second p-type semiconductor layer 573 is the band of the valence band of the first p-type semiconductor layer 572. Implementation is in that it is larger than the edge level E _V1 and the Fermi level E _{F1 of} the first p-type semiconductor layer 572 is larger than the Fermi level E _{F2 of} the second p-type semiconductor layer 573. Different from the semiconductor electrode 220 in the second embodiment.

Next, when the conductor 574, the first p-type semiconductor layer 572, and the second p-type semiconductor layer 573 are joined to each other, as shown in FIG. 30, the first p-type semiconductor layer 572 and the second p-type semiconductor layer 572 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 573, the band edge is bent. At this time, the band edge level E _V2 of the valence band of the second p-type semiconductor layer 573 is larger than the band edge level E _V1 of the valence band of the first p-type semiconductor layer 572, and the first Since the Fermi level E _F1 of the p-type semiconductor layer 572 is larger than the Fermi level E _{F2 of} the second p-type semiconductor layer 573, the band of the valence band is similar to the semiconductor electrode 220 in the second embodiment. At the edge, no Schottky barrier occurs at the junction surface between the first p-type semiconductor layer 572 and the second p-type semiconductor layer 573. However, as shown in FIG. 30, the band edge level E _V1 of the valence band of the first p-type semiconductor layer 572 is higher than the band edge level E _V2 of the valence band of the second p-type semiconductor layer 573. Becomes smaller.

At the junction surface between the first p-type semiconductor layer 572 and the conductor 574, the carrier moves so that the Fermi levels of the first p-type semiconductor layer 572 coincide with each other, so that the band edge near the junction surface of the first p-type semiconductor layer 572 is obtained. Bending occurs. At this time, since the Fermi level E _Fc of the conductor 574 is smaller than the Fermi level E _{F1 of} the first p-type semiconductor layer 572, similarly to the semiconductor electrode 220 in Embodiment 2, the conductor 574 and the second The junction of one p-type semiconductor layer 572 is in ohmic contact.

In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 220 in Embodiment 2, the band edge level E _V1 of the valence band of the first p-type semiconductor layer 572 is equal to that of the second p-type semiconductor layer 573. Since it becomes smaller than the band edge level E _V2 of the valence band, the movement of holes from the second p-type semiconductor layer 573 to the first p-type semiconductor layer 572 is prevented. Therefore, in the photoelectrochemical cell according to this comparative embodiment, the probability that electrons and holes generated in the second p-type semiconductor layer 573 are recombined by photoexcitation is higher than that in the photoelectrochemical cell 200 according to the second embodiment. As a result, the quantum efficiency of the hydrogen generation reaction due to light irradiation decreases.

<Comparative Form 2-6>
The photoelectrochemical cell which concerns on the comparison form 2-6 is demonstrated using FIG.31 and 32. FIG. FIG. 31 is a schematic diagram showing a band structure before bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment. FIG. 32 is a schematic diagram showing a band structure after bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment includes a substrate, a first p-type semiconductor layer 672 disposed on the substrate, a second p-type semiconductor layer 673 disposed on the first p-type semiconductor layer 672, and a conductive layer. And a body 674. As shown in FIG. 31, the semiconductor electrode of this comparative embodiment has a band edge level E _V2 of the valence band of the second p-type semiconductor layer 673 that is a band of the valence band of the first p-type semiconductor layer 672. The Fermi level E _{F1 of} the first p-type semiconductor layer 672 is greater than the edge level E _V1, greater than the Fermi level E _{F2 of} the second p-type semiconductor layer 673, and the Fermi level of the conductor 674 It differs from the semiconductor electrode 220 in Embodiment 2 in that the level E _Fc is larger than the Fermi level E _{F1 of} the first p-type semiconductor layer 672.

Next, when the conductor 674, the first p-type semiconductor layer 672, and the second p-type semiconductor layer 673 are joined to each other, as shown in FIG. 32, the first p-type semiconductor layer 672 and the second p-type semiconductor layer 672 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 673, bending of the band edge occurs. At this time, the band edge level E _V2 of the valence band of the second p-type semiconductor layer 673 is larger than the band edge level E _V1 of the valence band of the first p-type semiconductor layer 672, and the first Since the Fermi level E _F1 of the p-type semiconductor layer 672 is larger than the Fermi level E _{F2 of} the second p-type semiconductor layer 673, the band of the valence band is similar to the semiconductor electrode 220 in the second embodiment. At the edge, a Schottky barrier is not generated at the junction surface between the first p-type semiconductor layer 672 and the second p-type semiconductor layer 673. However, as shown in FIG. 32, the band edge level E _V1 of the valence band of the first p-type semiconductor layer 672 is higher than the band edge level E _V2 of the valence band of the second p-type semiconductor layer 673. Becomes smaller.

In addition, carriers move so that the Fermi levels of the first p-type semiconductor layer 672 and the conductor 674 coincide with each other. As a result, the band edge near the bonding surface of the first p-type semiconductor layer 672 is bent. Since the Fermi level E _Fc of the conductor 674 is larger than the Fermi level E _{F1 of} the first p-type semiconductor layer 672, the junction between the conductor 674 and the first p-type semiconductor layer 672 is a Schottky contact. Become.

In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 220 in Embodiment 2, the band edge level E _V1 of the valence band of the first p-type semiconductor layer 672 is equal to that of the second p-type semiconductor layer 673. Since it becomes smaller than the band edge level E _V2 of the valence band, the movement of holes from the second p-type semiconductor layer 673 to the first p-type semiconductor layer 672 is prevented. Further, in the case of this comparative embodiment, a Schottky barrier is generated at the junction surface between the conductor 674 and the first p-type semiconductor layer 672. This Schottky barrier prevents the movement of holes from the first p-type semiconductor layer 672 to the conductor 674. Therefore, in the photoelectrochemical cell according to this comparative embodiment, the probability that electrons and holes generated in the second p-type semiconductor layer 673 are recombined by photoexcitation is higher than that in the photoelectrochemical cell 200 according to the second embodiment. As a result, the quantum efficiency of the hydrogen generation reaction due to light irradiation decreases.

<Comparative Form 2-7>
The photoelectrochemical cell which concerns on the comparison form 2-7 is demonstrated using FIG. 33 and 34. FIG. FIG. 33 is a schematic diagram showing a band structure before bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment. FIG. 34 is a schematic diagram showing a band structure after bonding of the conductor, the first p-type semiconductor layer, and the second p-type semiconductor layer in this comparative embodiment.

The semiconductor electrode of this comparative embodiment is composed of a conductor 774, a first p-type semiconductor layer 772, and a second p-type semiconductor layer 773. As shown in FIG. 33, the semiconductor electrode of this comparative embodiment is different from that of Embodiment 2 in that the Fermi level E _Fc of the conductor 274 is larger than the Fermi level E _{F1 of} the first p-type semiconductor layer 772. Different from the semiconductor electrode 220.

Next, when the conductor 774, the first p-type semiconductor layer 772, and the second p-type semiconductor layer 773 are joined together, as shown in FIG. 34, the first p-type semiconductor layer 772 and the second p-type semiconductor layer 772 are joined. When the carriers move so that their Fermi levels coincide with each other at the joint surface with the type semiconductor layer 773, bending of the band edge occurs. The band edge level E _C2 of the conduction band of the second p-type semiconductor layer 773 and the band edge level E _V2 of the valence band are respectively the band edge level E of the conduction band of the first p-type semiconductor layer 772. _C1 and the band edge level E _V1 of the valence band, and the Fermi level E _{F1 of} the first p-type semiconductor layer 772 is smaller than the Fermi level E _{F2 of} the second p-type semiconductor layer 773. Therefore, no Schottky barrier is generated at the junction surface between the first p-type semiconductor layer 772 and the second p-type semiconductor layer 773.

At the junction surface between the first p-type semiconductor layer 772 and the conductor 774, carriers move so that their Fermi levels coincide with each other, whereby a band edge in the vicinity of the junction surface of the first p-type semiconductor layer 772 is obtained. Bending occurs. Since the Fermi level E _Fc of the conductor 774 is larger than the Fermi level E _{F1 of} the first p-type semiconductor layer 772, the junction between the conductor 774 and the first p-type semiconductor layer 772 is a Schottky contact. Become.

  In the case of the semiconductor electrode of this comparative embodiment, unlike the semiconductor electrode 220 in Embodiment 2, a Schottky barrier is generated at the junction surface between the conductor 774 and the first p-type semiconductor layer 772. This Schottky barrier prevents the movement of holes from the first p-type semiconductor layer 772 to the conductor 774. Therefore, in the photoelectrochemical cell of this comparative embodiment, the probability that electrons and holes generated in the second p-type semiconductor layer 773 are recombined by photoexcitation is higher than that of the photoelectrochemical cell 200 of the second embodiment. The quantum efficiency of the hydrogen generation reaction due to light irradiation is reduced.

(Embodiment 3)
The configuration of the photoelectrochemical cell according to Embodiment 3 of the present invention will be described with reference to FIG. FIG. 35 is a schematic diagram showing the configuration of the photoelectrochemical cell of the present embodiment.

  In the photoelectrochemical cell 300 according to the present embodiment, the semiconductor electrode 320 is disposed on the substrate 321, the first n-type semiconductor layer 322 disposed on the substrate 321, and the first n-type semiconductor layer 322. The second n-type semiconductor layer 323 and the conductor 324 are included. On the other hand, the counter electrode 330 is disposed on the conductor 324 (on the surface of the conductor 324 opposite to the surface on which the first n-type semiconductor layer 322 is disposed). Note that the substrate 321, the first n-type semiconductor layer 322, the second n-type semiconductor layer 323, and the conductor 324 are each formed using the substrate 121, the first n-type semiconductor layer 122, and the first conductor 324 in Embodiment Mode 1, respectively. 2 is the same as the n-type semiconductor layer 123 and the conductor 124.

  According to the configuration in which the counter electrode 330 is disposed on the conductor 324 as in the present embodiment, a conducting wire for electrically connecting the semiconductor electrode 320 and the counter electrode 330 is not necessary. As a result, the resistance loss due to the conducting wire is eliminated, so that the quantum efficiency of the hydrogen generation reaction by light irradiation can be further improved. Moreover, according to such a structure, the semiconductor electrode 320 and the counter electrode 330 can be electrically connected by a simple process. Note that in this embodiment, the above-described structure in which the counter electrode is disposed on the conductor is applied to the photoelectrochemical cell described in Embodiment 1, and such a structure is described in Embodiment 2. It can also be applied to photoelectrochemical cells.

(Embodiment 4)
The structure of the photoelectrochemical cell of Embodiment 4 of this invention is demonstrated using FIG. FIG. 36 is a schematic diagram showing the configuration of the photoelectrochemical cell of the present embodiment.

  As shown in FIG. 36, the photoelectrochemical cell 400 of this embodiment includes a housing (container) 410, a semiconductor electrode 420, a counter electrode 430, and a separator 460. The interior of the housing 410 is separated into two chambers, a first chamber 470 and a second chamber 480, by a separator 460. In the first chamber 470 and the second chamber 480, an electrolytic solution 440 containing water is accommodated, respectively.

  A semiconductor electrode 420 is disposed in the first chamber 470 at a position in contact with the electrolytic solution 440. The semiconductor electrode 420 includes a substrate 421, a first n-type semiconductor layer 422 disposed on the substrate 421, a second n-type semiconductor layer 423 disposed on the first n-type semiconductor layer 422, and a conductor. 424. The first chamber 470 includes a first exhaust port 471 for exhausting oxygen generated in the first chamber 470 and a water supply port 472 for supplying water into the first chamber 470. . A portion of the housing 410 facing the second n-type semiconductor layer 423 of the semiconductor electrode 420 disposed in the first chamber 470 (hereinafter abbreviated as a light incident portion 410a) emits light such as sunlight. It is composed of a material to be transmitted.

  On the other hand, a counter electrode 430 is disposed in the second chamber 480 at a position in contact with the electrolytic solution 440. In addition, the second chamber 480 includes a second exhaust port 481 for exhausting hydrogen generated in the second chamber 480.

  The conductor 424 and the counter electrode 430 in the semiconductor electrode 420 are electrically connected by a conducting wire 450.

  The substrate 421, the first n-type semiconductor layer 422, the second n-type semiconductor layer 423, and the conductor 424 of the semiconductor electrode 420 in this embodiment are the same as the substrate 121 of the semiconductor electrode 120 in the first embodiment, the first The n-type semiconductor layer 122, the second n-type semiconductor layer 123, and the conductor 124 have the same configuration. Therefore, the semiconductor electrode 420 has the same effects as the semiconductor electrode 120 of the first embodiment. The counter electrode 430 and the electrolytic solution 440 are the same as the counter electrode 130 and the electrolytic solution 140 in the first embodiment, respectively.

  The separator 460 is formed of a material having a function of permeating the electrolytic solution 440 and blocking each gas generated in the first chamber 470 and the second chamber 480. Examples of the material of the separator 460 include a solid electrolyte such as a polymer solid electrolyte. Examples of the polymer solid electrolyte include an ion exchange membrane such as Nafion (registered trademark). Using such a separator, the internal space of the container is divided into two regions, and the electrolyte solution and the surface of the semiconductor electrode (the surface of the second n-type semiconductor layer) are brought into contact with each other in one region, By adopting a configuration in which the electrolytic solution and the surface of the counter electrode are brought into contact with each other, oxygen and hydrogen generated inside the container can be easily separated.

  Note that in this embodiment mode, the photoelectrochemical cell 400 using the semiconductor electrode 420 having the same configuration as the semiconductor electrode 120 in Embodiment Mode 1 is described, but the semiconductor electrode in Embodiment Mode 2 is used instead of the semiconductor electrode 420. It is also possible to use the electrode 220.

(Embodiment 5)
The configuration of the energy system according to the fifth embodiment of the present invention will be described with reference to FIG. FIG. 37 is a schematic diagram showing the configuration of the energy system of the present embodiment.

  As shown in FIG. 37, the energy system 500 of the present embodiment includes a photoelectrochemical cell 400, a hydrogen reservoir 510, a fuel cell 520, and a storage battery 530.

  The photoelectrochemical cell 400 is the photoelectrochemical cell described in Embodiment 4, and its specific configuration is as shown in FIG. Therefore, detailed description is omitted here.

  The hydrogen storage 510 is connected to the second chamber 480 (see FIG. 36) of the photoelectrochemical cell 400 by the first pipe 541. As the hydrogen storage 510, for example, a compressor that compresses hydrogen generated in the photoelectrochemical cell 400 and a high-pressure hydrogen cylinder that stores hydrogen compressed by the compressor can be used.

  The fuel cell 520 includes a power generation unit 521 and a fuel cell control unit 522 for controlling the power generation unit 521. The fuel cell 520 is connected to the hydrogen storage 510 through a second pipe 542. The second pipe 542 is provided with a cutoff valve 543. As the fuel cell 520, for example, a solid polymer electrolyte fuel cell can be used.

  The positive electrode and the negative electrode of the storage battery 530 are electrically connected to the positive electrode and the negative electrode of the power generation unit 521 in the fuel cell 520 by the first wiring 544 and the second wiring 545, respectively. The storage battery 530 is provided with a capacity measurement unit 546 for measuring the remaining capacity of the storage battery 530. As the storage battery 530, for example, a lithium ion battery can be used.

  Next, the operation of the energy system 500 of the present embodiment will be described with reference to FIGS. 36 and 37.

  When the surface of the second n-type semiconductor layer 423 of the semiconductor electrode 420 disposed in the first chamber 470 is irradiated with sunlight through the light incident portion 410a of the photoelectrochemical cell 400, the second n-type semiconductor Electrons and holes are generated in the layer 423. The holes generated at this time move to the surface side of the second n-type semiconductor layer 423. Thereby, on the surface of the second n-type semiconductor layer 423, water is decomposed by the reaction formula (1) to generate oxygen.

  On the other hand, electrons are bent at the band edge of the conduction band at the interface between the second n-type semiconductor layer 423 and the first n-type semiconductor layer 422 and at the interface between the first n-type semiconductor layer 422 and the conductor 424. And move to the conductor 424. The electrons that have moved to the conductor 424 move to the counter electrode 430 side that is electrically connected to the conductor 424 through the conductive wire 450. Thereby, hydrogen is generated on the surface of the counter electrode 430 according to the reaction formula (2).

  At this time, similarly to the semiconductor electrode 120 in Embodiment 1, since no Schottky barrier is generated at the junction surface between the second n-type semiconductor layer 423 and the first n-type semiconductor layer 422, electrons are prevented. Without being moved from the second n-type semiconductor layer 423 to the first n-type semiconductor layer 422. Further, since no Schottky barrier is generated at the bonding surface between the first n-type semiconductor layer 422 and the conductor 424, electrons can move to the conductor 424 without being blocked. Therefore, the probability of recombination of electrons and holes generated in the first n-type semiconductor layer 423 by photoexcitation is reduced, and the quantum efficiency of the hydrogen generation reaction by light irradiation can be improved.

  Oxygen generated in the first chamber 470 is exhausted out of the photoelectrochemical cell 400 through the first exhaust port 471. On the other hand, hydrogen generated in the second chamber 480 is supplied into the hydrogen reservoir 510 through the second exhaust port 481 and the first pipe 541.

  When generating power in the fuel cell 520, the shutoff valve 543 is opened by a signal from the fuel cell control unit 522, and the hydrogen stored in the hydrogen storage 510 is transferred to the power generation unit 521 of the fuel cell 520 by the second pipe 542. Supplied.

  Electricity generated in the power generation unit 521 of the fuel cell 520 is stored in the storage battery 530 via the first wiring 544 and the second wiring 545. Electricity stored in the storage battery 530 is supplied to homes, businesses, and the like by the third wiring 547 and the fourth wiring 548.

  According to the photoelectrochemical cell 400 in the present embodiment, it is possible to improve the quantum efficiency of the hydrogen generation reaction by light irradiation. Therefore, according to the energy system 500 of this Embodiment provided with such a photoelectrochemical cell 400, electric power can be supplied efficiently.

  In the present embodiment, an example of an energy system using the photoelectrochemical cell 400 described in the fourth embodiment is shown. However, the photoelectrochemical cells 100, 200, and 300 described in the first to third embodiments are illustrated. The same effect can be obtained by using.

  Examples of the present invention will be specifically described below.

Example 1
As Example 1, a photoelectrochemical cell having the same configuration as the photoelectrochemical cell 100 shown in FIG. Hereinafter, the photoelectrochemical cell of Example 1 will be described with reference to FIG.

  The photoelectrochemical cell 100 of Example 1 was provided with a square glass container (container 110) having an opening at the top, a semiconductor electrode 120, and a counter electrode. In the glass container 110, 1 mol / L NaOH aqueous solution was accommodated as the electrolyte solution 140.

  The semiconductor electrode 120 was produced by the following procedure.

  First, a 2-inch square sapphire substrate was prepared. On this sapphire substrate, an intrinsic GaN (i-GaN) crystal film with a thickness of 30 nm was formed by MOCVD (metalorganic chemical vapor deposition) using trimethylgallium as a reaction gas. This sapphire substrate with an i-GaN crystal film is used as the substrate 121, and an n-GaN crystal film (first n-type semiconductor layer 122) is formed thereon by MOCVD using trimethylgallium and doped silane gas as reaction gases. Was formed with a thickness of 2000 nm. Further, an n-GaInN crystal film (second n-type semiconductor layer 123) was formed to a thickness of 100 nm by MOCVD using triethylgallium, trimethylindium, and doped silane gas as reaction gases.

  A laminate in which an i-GaN crystal film, an n-GaN crystal film, and an n-GaInN crystal film are provided on a sapphire substrate is cut into a 2 cm × 1 cm square, and a 0.9 cm × 1 cm portion of the n-GaInN crystal is included. The film (thickness 100 nm) was removed by etching. An Al film (thickness: 100 nm) was vapor-deposited on the portion where the n-GaInN crystal film was removed to form a conductor 124. Note that a gap of 0.1 cm was provided so that the n-GaInN crystal film as the second n-type semiconductor layer 123 did not contact the Al film as the conductor 124. Thus, the semiconductor electrode 120 of Example 1 was produced.

  The semiconductor electrode 120 produced as described above was placed in the glass container 110 so that the surface of the second n-type semiconductor layer 123 faces the light incident part 110a of the glass container 110.

  A platinum plate was used as the counter electrode 130. The conductor 124 of the semiconductor electrode 120 and the counter electrode 130 were electrically connected by a conducting wire 150.

About the photoelectrochemical cell 100 of Example 1 produced in this way, the pseudo-sunlight irradiation experiment was conducted. In the pseudo-sunlight irradiation experiment, a solar simulator manufactured by Celic Corporation is used as the pseudo-sunlight, and the strength of the surface of the second n-type semiconductor layer 123 in the semiconductor electrode 120 of the photoelectrochemical cell 100 via the light incident part 110a. Irradiation with light of 1 kW / m ² was performed. The gas generated on the surface of the counter electrode 130 was collected for 30 minutes, and the component analysis of the collected gas and the generation amount were measured by gas chromatography. Further, the photocurrent flowing between the semiconductor electrode 120 and the counter electrode 130 was measured with an ammeter. The apparent quantum efficiency was determined using the amount of gas produced at the counter electrode 130.

As a result of analyzing the gas collected in the photoelectrochemical cell of Example 1, it was confirmed that hydrogen was generated on the counter electrode. The production rate of hydrogen was 1.9 × 10 ⁻⁷ L / s. Further, since the photocurrent flowing between the semiconductor electrode and the counter electrode was 1.5 mA / cm ² , it was confirmed that water was electrolyzed stoichiometrically. The apparent quantum efficiency was calculated by using the following calculation formula and found to be about 63%.

Apparent quantum efficiency = {(observed photocurrent density [mA / cm ² ]) / (photocurrent density that can be generated by sunlight that can be absorbed in the band gap of the semiconductor material used for the second n-type semiconductor layer) [MA / cm ² ])} × 100

In Example 1, the observed photocurrent density is 1.5 mA / cm ² , and the band gap of the semiconductor material used for the second n-type semiconductor layer (Ga _0.9 In _0.1 N band gap (2. The photocurrent density that can be generated by sunlight that can be absorbed at 9 eV)) was 2.4 mA / cm ² .

(Example 2)
As Example 2, a photoelectrochemical cell having the same structure as the photoelectrochemical cell 100 shown in FIG. In Example 2, a photoelectrochemical cell in which only the material of the semiconductor electrode 120 was different from that in Example 1 was produced. The semiconductor electrode 120 of Example 2 was produced as follows.

  First, a 2 cm × 1 cm square glass substrate was prepared as the substrate 121. On this glass substrate, a 500 nm-thick titanium oxide film (anater polycrystal) was formed as the first n-type semiconductor layer 122 by sputtering. A cadmium sulfide film having a thickness of 1 μm was formed as a second n-type semiconductor layer 123 on the first n-type semiconductor layer 122 by a chemical precipitation method using cadmium acetate and thiourea. Of this, the cadmium sulfide film (thickness 1 μm) of 0.9 cm × 1 cm was removed by etching. A Ti film (thickness 150 nm) was vapor-deposited on the portion where the cadmium sulfide film was removed to form a conductor 124. Note that a gap of 0.1 cm was provided so that the cadmium sulfide film as the second n-type semiconductor layer 123 did not contact the Ti film as the conductor 124. Thus, the semiconductor electrode 120 of Example 2 was produced.

Thus the fabricated photoelectrochemical cell 100 of Embodiment 2, the glass container 110, as the electrolytic solution 140, Na ₂ SO ₃ aqueous solution 0.01 mol / L containing Na ₂ S of 0.01 mol / L A pseudo-sunlight irradiation experiment was carried out in the same manner as in Example 1 except that. As a result of analyzing the gas collected in the photoelectrochemical cell of Example 2, it was confirmed that hydrogen was generated on the counter electrode. The production rate of hydrogen was 2.9 × 10 ⁻⁷ L / s. In addition, since the photocurrent flowing between the semiconductor electrode and the counter electrode was 2.3 mA / cm ² , it was confirmed that water was electrolyzed quantitatively. Moreover, the apparent quantum efficiency was calculated | required using the said calculation formula shown in Example 1. FIG. In this example, the observed photocurrent density is 2.3 mA / cm ^2, which is the band gap of the semiconductor material used for the second n-type semiconductor layer (CdS band gap (2.5 eV)). The photocurrent density that can be generated by sunlight that can be absorbed was 6.5 mA / cm ² . As a result, the apparent quantum efficiency of Example 2 was calculated to be about 35%.

In the photoelectrochemical cell of Example 2, as the electrolytic solution, since it uses the aqueous Na ₂ SO ₃ solution containing Na ₂ S, when irradiated with light, the semiconductor electrode, the above reaction formula ( It is considered that the reaction shown in the following reaction formula (3) is proceeding instead of the oxygen generation reaction by 1). In addition, it is thought that reaction shown in the above-mentioned reaction formula (2) is progressing in the counter electrode.

2h ⁺ + S ^2- → S (3)

(Example 3)
As Example 3, a photoelectrochemical cell having the same structure as the photoelectrochemical cell 100 shown in FIG. In Example 3, a photoelectrochemical cell in which only the material of the semiconductor electrode 120 was different from that in Example 1 was produced. The semiconductor electrode 120 of Example 3 was produced as follows.

First, a 2 cm × 1 cm square glass substrate was prepared as the substrate 121. On this glass substrate, as the first n-type semiconductor layer 122, InGaZn ₂ O ₅ is used as a target, the oxygen partial pressure of the chamber is 0.1 Pa, the substrate temperature is room temperature, and a 150 nm-thick InGaZn ₂ O film is formed by sputtering. _Five films were formed as the first n-type semiconductor layer 122. After a 0.9 cm × 1 cm portion on the first n-type semiconductor layer 122 is masked, an InGaZn layer is formed as a second n-type semiconductor layer 123 on the unmasked portion of the first n-type semiconductor layer 122. An InGaZn ₂ O ₅ film having a thickness of 500 nm and a portion of oxygen substituted with nitrogen was formed by a reactive sputtering method with a ₂ O ₅ target, a nitrogen pressure of the chamber of 0.1 Pa, and a substrate temperature of room temperature. Further, the mask was removed, and an Ag film (thickness 150 nm) was deposited on the masked portion on the first n-type semiconductor layer 122 to obtain a conductor 124. Note that a gap of 0.1 cm was provided so that the second n-type semiconductor layer 123 and the conductor 124 did not contact each other. Thus, the semiconductor electrode 120 of Example 3 was produced.

About the photoelectrochemical cell 100 of Example 3 produced in this way, Example 1 except that 0.01 mol / L Na ₂ CO ₃ aqueous solution is accommodated in the glass container 110 as the electrolytic solution 140. Pseudo sunlight irradiation experiment was conducted by the same method. As a result of analyzing the gas collected in the photoelectrochemical cell of Example 3, it was confirmed that hydrogen was generated on the counter electrode. The production rate of hydrogen was 2.5 × 10 ⁻⁷ L / s. Further, since the photocurrent flowing between the semiconductor electrode and the counter electrode was 2.0 mA / cm ² , it was confirmed that water was electrolyzed stoichiometrically. Moreover, the apparent quantum efficiency was calculated | required using the said calculation formula shown in Example 1. FIG. In this example, the observed photocurrent density was 2.0 mA / cm ² . The band gap of the InGaZn ₂ O ₅ film used for the second n-type semiconductor layer 123 in which part of oxygen is replaced with nitrogen is about 1.8 eV from the UV-Vis absorption spectrum. The photocurrent density that can be generated by sunlight that can be absorbed was 19.7 mA / cm ² . As a result, the apparent quantum efficiency of Example 3 was calculated to be about 10%.

Example 4
As Example 4, a photoelectrochemical cell having the same configuration as the photoelectrochemical cell 300 shown in FIG. 35 was produced. Hereinafter, the photoelectrochemical cell of Example 4 will be described with reference to FIG.

  The photoelectrochemical cell 300 of Example 4 was provided with a square glass container (container 110) having an opening at the top, a semiconductor electrode 320, and a counter electrode 330. In the glass container 110, 1 mol / L NaOH aqueous solution was accommodated as the electrolyte solution 140.

  The semiconductor electrode 320 was produced by the following procedure.

  First, a 2-inch square sapphire substrate was prepared. On this sapphire substrate, an intrinsic GaN (i-GaN) crystal film was formed with a thickness of 30 nm by MOCVD using trimethylgallium as a reaction gas. This sapphire substrate with an i-GaN crystal film is used as a substrate 321, and an n-GaN crystal film (first n-type semiconductor layer 322) is formed thereon by MOCVD using trimethylgallium and a doped silane gas as reaction gases. Was formed with a thickness of 2000 nm. Further, an n-GaInN crystal film (second n-type semiconductor layer 323) was formed with a thickness of 100 nm by MOCVD using triethylgallium, trimethylindium, and doped silane gas as reaction gases.

  A laminate in which an i-GaN crystal film, an n-GaN crystal film, and an n-GaInN crystal film are provided on a sapphire substrate is cut into a 2 cm × 1 cm square, and a 0.9 cm × 1 cm portion of the n-GaInN crystal is included. The film (thickness 100 nm) was removed by etching. An Al film (thickness: 100 nm) was deposited on the portion from which the n-GaInN crystal film was removed, whereby a conductor 324 was obtained. Note that a gap of 0.1 cm was provided so that the n-GaInN crystal film as the second n-type semiconductor layer 323 and the Al film as the conductor 324 were not in contact with each other. Thus, the semiconductor electrode 320 of Example 4 was produced.

  A Pt film having a thickness of 10 nm was formed as a counter electrode 330 on the conductor 324 of the semiconductor electrode 320 by sputtering.

  The semiconductor electrode 320 and the counter electrode 330 produced as described above were arranged in the glass container 110 so that the surface of the second n-type semiconductor layer 323 faces the light incident part 110a of the glass container 110.

With respect to the photoelectrochemical cell 300 of Example 4 manufactured in this way, a pseudo-sunlight irradiation experiment was performed in the same manner as in Example 1. As a result of analyzing the gas collected in the photoelectrochemical cell of Example 4, it was confirmed that hydrogen was generated on the counter electrode. The production rate of hydrogen was 2.1 × 10 ⁻⁷ L / s. Further, since the photocurrent flowing between the semiconductor electrode and the counter electrode was 1.7 mA / cm ² , it was confirmed that water was electrolyzed quantitatively. Moreover, the apparent quantum efficiency was calculated | required using the said calculation formula shown in Example 1. FIG. In this example, the observed photocurrent density is 1.7 mA / cm ² , and the band gap of the semiconductor material used for the second n-type semiconductor layer (Ga _0.9 In _0.1 N band gap (2. The photocurrent density that can be generated by sunlight that can be absorbed at 9 eV)) was 2.4 mA / cm ² . As a result, the apparent quantum efficiency of Example 4 was calculated to be about 71%.

(Comparative Example 1)
As Comparative Example 1, a photoelectrochemical cell in which only the first n-type semiconductor layer was different from Example 1 was produced. The first n-type semiconductor layer of Comparative Example 1 was formed by reducing the amount of doped silane gas as compared with Example 1.

For the photoelectrochemical cell of Comparative Example 1 produced in this way, a pseudo-sunlight irradiation experiment was performed in the same manner as in Example 1. As a result of analyzing the gas collected in the photoelectrochemical cell of Comparative Example 1, it was confirmed that hydrogen was generated on the counter electrode. The production rate of hydrogen was 6.7 × 10 ⁻⁸ L / s. In addition, since the photocurrent flowing between the semiconductor electrode and the counter electrode was 0.52 mA / cm ² , it was confirmed that water was electrolyzed quantitatively. Moreover, the apparent quantum efficiency was calculated | required using the said calculation formula shown in Example 1. FIG. In this comparative example, the observed photocurrent density was 0.52 mA / cm ² , and the band gap of the semiconductor material used for the second n-type semiconductor layer (Ga _0.9 In _0.1 N band gap (2. The photocurrent density that can be generated by sunlight that can be absorbed at 9 eV)) was 2.4 mA / cm ² . As a result, the apparent quantum efficiency of Comparative Example 1 was calculated to be about 22%.

(Comparative Example 2)
As Comparative Example 2, a photoelectrochemical cell using a semiconductor electrode not provided with the first n-type semiconductor layer was produced. An n-GaInN crystal film (thickness: 100 nm), which is a second n-type semiconductor layer, is formed on a substrate (a sapphire substrate with an i-GaN crystal film) prepared in the same manner as in Example 1, as in Example 1. It formed by the method of. The obtained laminated body was cut into 2 cm × 1 cm, and an Al film (thickness: 100 nm) was deposited as a conductor on the n-GaInN crystal film in a region of 0.9 cm × 1 cm. The semiconductor electrode was made. The configuration other than the semiconductor electrode was the same as in Example 1.

For the photoelectrochemical cell of Comparative Example 2 produced in this way, a pseudo-sunlight irradiation experiment was performed in the same manner as in Example 1. As a result of analyzing the gas collected in the photoelectrochemical cell of Comparative Example 2, it was confirmed that hydrogen was generated on the counter electrode. The production rate of hydrogen was 7.8 × 10 ⁻⁸ L / s. Further, since the photocurrent flowing between the semiconductor electrode and the counter electrode was 0.6 mA / cm ² , it was confirmed that water was electrolyzed stoichiometrically. Moreover, the apparent quantum efficiency was calculated | required using the said calculation formula shown in Example 1. FIG. In this comparative example, the observed photocurrent density is 0.6 mA / cm ² , and the band gap of the semiconductor material used for the second n-type semiconductor layer (the band gap of Ga _0.9 In _0.1 N (2. The photocurrent density that can be generated by sunlight that can be absorbed at 9 eV)) was 2.4 mA / cm ² . As a result, the apparent quantum efficiency of Comparative Example 2 was calculated to be about 25%.

(Comparative Example 3)
As Comparative Example 3, a photoelectrochemical cell using a semiconductor electrode not provided with the first n-type semiconductor layer was produced. The same glass substrate as that used in Example 3 was prepared, and an InGaZn ₂ O ₅ film in which a part of oxygen was replaced with nitrogen used as the second n-type semiconductor layer in Example 3 (Thickness 500 nm) was formed in the same manner as in Example 3. The obtained laminated body was cut into 2 cm × 1 cm, and an Ag film (thickness: 100 nm) was deposited as a conductor on the InGaZn ₂ O ₅ film in a region of 0.9 cm × 1 cm. The semiconductor electrode was made. The configuration other than the semiconductor electrode was the same as in Example 3.

With respect to the photoelectrochemical cell of Comparative Example 3 produced in this manner, a pseudo-sunlight irradiation experiment was performed in the same manner as in Example 3. As a result of analyzing the gas collected in the photoelectrochemical cell of Comparative Example 3, it was confirmed that hydrogen was generated on the counter electrode. The production rate of hydrogen was 1.0 × 10 ⁻⁷ L / s. Further, since the photocurrent flowing between the semiconductor electrode and the counter electrode was 0.8 mA / cm ² , it was confirmed that water was electrolyzed stoichiometrically. Moreover, the apparent quantum efficiency was calculated | required using the said calculation formula shown in Example 1. FIG. In this comparative example, the observed photocurrent density was 0.8 mA / cm ² , and the InGaZn ₂ O ₅ film used as the second n-type semiconductor layer in which part of oxygen was replaced with nitrogen The band gap was about 1.8 eV from the UV-Vis absorption spectrum, and the photocurrent density that could be generated by sunlight that could be absorbed by the band gap was 19.7 mA / cm ² . Thereby, the apparent quantum efficiency of Comparative Example 3 was calculated to be about 4%.

Conductor, material of first n-type semiconductor layer and second n-type semiconductor layer, Fermi level E _F constituting semiconductor electrode in each photoelectrochemical cell of Examples 1-4 and Comparative Examples 1-3 Table 1 shows (unit: eV), band edge level E _C (unit: eV) of the conduction band, and band edge level E _V (unit: eV) of the valence band. Note that the Fermi level, the band edge level of the conduction band, and the band edge level of the valence band are values based on the vacuum level when in contact with an electrolyte having a pH of 0 and a temperature of 25 ° C. The Fermi level was obtained from measurement of a flat band potential. The band edge level of the conduction band and the band edge level of the valence band are cited from the literature. The quantum efficiencies of the photoelectrochemical cells of Examples 1 to 4 and Comparative Examples 1 to 3 are also shown in Table 1.

As can be seen from Table 1, the semiconductor electrode in each photoelectrochemical cell according to Examples 1 to 4 has the band edge level E _{C2 in} the conduction band of the second n-type semiconductor and the band edge level E in the valence band. _V2 is larger than the band edge level E _C1 of the conduction band of the first n-type semiconductor and the band edge level E _V1 of the valence band, respectively, and the magnitude relation of the Fermi level is the second n-type semiconductor. The relationship of layer <first n-type semiconductor layer <conductor is satisfied. Further, in each of the semiconductor electrodes of Examples 1 to 4, the Fermi level of the first n-type semiconductor layer is −4.44 eV or more in a state where it is in contact with the electrolyte solution having a pH value of 0 and a temperature of 25 ° C., and The band edge level of the valence band of the second n-type semiconductor layer was −5.67 eV or less.

  As described above, the conductors, the first n-type semiconductor layer, and the second n-type semiconductor layer in Examples 1 to 4 are the conductor 124, the first n-type semiconductor layer 122, and the first n-type semiconductor layer in Embodiment 1. 2 had the same band structure as that of the n-type semiconductor layer 123 (see FIGS. 2 and 3).

The semiconductor electrode in the photoelectrochemical cell according to Comparative Example 1 has a band edge level E _{C2 in} the conduction band and a band edge level E _{V2 in the} valence band of the second n-type semiconductor layer, respectively. Although it is larger than the band edge level E _C1 of the conduction band of the n-type semiconductor layer and the band edge level E _V1 of the valence band, the magnitude relationship of the Fermi level is as follows: second n-type semiconductor layer> first n-type The semiconductor layer was <conductor. In addition, when the semiconductor electrode is in contact with the electrolytic solution having a pH value of 0 and a temperature of 25 ° C., the band edge level of the valence band of the second n-type semiconductor layer is −5.67 eV or less, The n-type semiconductor layer had a Fermi level of less than -4.44 eV.

  As described above, the conductor, the first n-type semiconductor layer, and the second n-type semiconductor layer in Comparative Example 1 are the conductor 164 and the first n-type semiconductor in Comparative Example 1-1 in Embodiment 1. The band structure similar to that of the layer 162 and the second n-type semiconductor layer 163 (see FIGS. 4 and 5) was obtained.

  Although the material systems of the first n-type semiconductor layer and the second n-type semiconductor layer are the same, the quantum levels of Example 1 and Comparative Example 1 having different energy level relationships (here, the Fermi level magnitude relationship) are different. When comparing the efficiency, the photoelectrochemical cell of Example 1 that satisfies the band structure in the present invention has a higher quantum efficiency than that of Comparative Example 1. Further, when the second n-type semiconductor layer is made of the same material but the first and second n-type semiconductor layers are different in configuration in terms of the presence or absence of the first n-type semiconductor layer, the first n-type semiconductor is compared. The quantum efficiency of Example 1 in which the layer was provided was about twice as high as that of Comparative Example 2 in which the first n-type semiconductor layer was not provided.

  In the case of Example 2 satisfying the structure of the photoelectrochemical cell of the present invention, good quantum efficiency was obtained as in Example 1. However, Example 1 in which n-GaN is used for the first n-type semiconductor layer and n-type Group III nitride semiconductor is used for the second n-type semiconductor layer can achieve higher quantum efficiency. did it.

  Although the material of the second n-type semiconductor layer is the same, the comparison between Example 3 and Comparative Example 3 that differ in configuration in the presence or absence of the first n-type semiconductor layer reveals that the first n-type semiconductor layer is The quantum efficiency of the provided Example 3 was twice or more higher than the quantum efficiency of the Comparative Example 3 in which the first n-type semiconductor layer was not provided.

  Further, when the results of Example 1 and Example 4 in which the configuration of the semiconductor electrode is the same but the positional relationship of the counter electrodes are different from each other are compared, the apparent quantum efficiency of Example 1 is 63%. The apparent quantum efficiency of Example 4 was 71%. From this result, in the photoelectrochemical cell of Example 4, since the resistance loss resulting from a conducting wire was lost, it was confirmed that the apparent quantum efficiency was further improved as compared with the photoelectrochemical cell of Example 1.

  In Examples 1 to 4, an example in which an n-type semiconductor layer is used as a semiconductor electrode is shown. However, similar results can be obtained even if a p-type semiconductor layer is used instead of an n-type semiconductor layer. For example, in Example 1, instead of the first n-type semiconductor layer, a p-GaN film is formed by MOCVD as the first p-type semiconductor layer, and the second p-type semiconductor layer is formed on the first p-type semiconductor layer. A p-type semiconductor layer may be formed by forming a p-GaInN film by MOCVD as a p-type semiconductor layer. In this case, as in the case of Example 1, it is estimated that the apparent quantum efficiency is improved.

  According to the photoelectrochemical cell and the energy system of the present invention, the quantum efficiency of the hydrogen generation reaction by light irradiation can be improved, so that it is useful as a household power generation system or the like.

Claims (14)

A substrate, a first n-type semiconductor layer disposed on the substrate, and a second n-type semiconductor layer and a conductor disposed on the first n-type semiconductor layer so as to be spaced apart from each other; Including a semiconductor electrode;
A counter electrode electrically connected to the conductor;
An electrolyte in contact with the surface of the second n-type semiconductor layer and the counter electrode;
A container containing the semiconductor electrode, the counter electrode and the electrolyte;
With
In the semiconductor electrode, with reference to the vacuum level,
(I) The band edge levels of the conduction band and the valence band in the second n-type semiconductor layer are larger than the band edge levels of the conduction band and the valence band in the first n-type semiconductor layer, respectively. ,
(II) The Fermi level of the first n-type semiconductor layer is larger than the Fermi level of the second n-type semiconductor layer, and
(III) The Fermi level of the conductor is larger than the Fermi level of the first n-type semiconductor layer,
Generating hydrogen by irradiating the second n-type semiconductor layer with light;
Photoelectrochemical cell. When the pH value of the electrolyte is 0 and the temperature is 25 ° C., the vacuum level is used as a reference,
The Fermi level of the first n-type semiconductor layer is −4.44 eV or more, and the band edge level of the valence band in the second n-type semiconductor layer is −5.67 eV or less. Item 2. The photoelectrochemical cell according to Item 1.   The photoelectrochemical cell according to claim 1, wherein the first n-type semiconductor layer is made of n-type gallium nitride.   2. The second n-type semiconductor layer is made of an n-type group III nitride semiconductor containing gallium and at least one element selected from the group consisting of indium and aluminum as a group III element. A photoelectrochemical cell according to 1. The substrate is a sapphire substrate;
The photoelectrochemical cell according to claim 1, wherein the first n-type semiconductor layer and the second n-type semiconductor layer are crystal films obtained by epitaxial growth.   The photoelectrochemical cell according to claim 1, wherein the first n-type semiconductor layer is made of an n-type oxide semiconductor containing gallium, indium, and zinc.   2. The photoelectrochemical cell according to claim 1, wherein the second n-type semiconductor layer is made of an n-type semiconductor having a composition in which part of oxygen of an oxide containing gallium, indium, and zinc is substituted with nitrogen. A substrate, a first p-type semiconductor layer disposed on the substrate, and a second p-type semiconductor layer and a conductor disposed on the first p-type semiconductor layer and spaced apart from each other; Including a semiconductor electrode;
A counter electrode electrically connected to the conductor;
An electrolyte in contact with the surface of the second p-type semiconductor layer and the counter electrode;
A container containing the semiconductor electrode, the counter electrode and the electrolyte;
With
In the semiconductor electrode, with reference to the vacuum level,
(I) The band edge levels of the conduction band and the valence band in the second p-type semiconductor layer are smaller than the band edge levels of the conduction band and the valence band in the first p-type semiconductor layer, respectively. ,
(II) the Fermi level of the first p-type semiconductor layer is smaller than the Fermi level of the second p-type semiconductor layer, and
(III) The Fermi level of the conductor is smaller than the Fermi level of the first p-type semiconductor layer,
Generating hydrogen by irradiating the second p-type semiconductor layer with light;
Photoelectrochemical cell. When the pH value of the electrolyte is 0 and the temperature is 25 ° C., the vacuum level is used as a reference,
The Fermi level of the first p-type semiconductor layer is −5.67 eV or less, and the band edge level of the conduction band in the second p-type semiconductor layer is −4.44 eV or more. 9. The photoelectrochemical cell according to 8.   The photoelectrochemical cell according to claim 8, wherein the first p-type semiconductor layer is made of p-type gallium nitride.   The second p-type semiconductor layer is made of a p-type group III nitride semiconductor containing gallium and at least one element selected from the group consisting of indium and aluminum as a group III element. A photoelectrochemical cell according to 1. The substrate is a sapphire substrate;
The photoelectrochemical cell according to claim 8, wherein the first p-type semiconductor layer and the second p-type semiconductor layer are crystal films obtained by epitaxial growth.   The photoelectrochemical cell according to claim 1 or 8, wherein the counter electrode is disposed on the conductor. The photoelectrochemical cell according to claim 1 or 8,
A hydrogen reservoir that is connected to the photoelectrochemical cell by a first pipe and stores hydrogen generated in the photoelectrochemical cell;
A fuel cell that is connected to the hydrogen reservoir by a second pipe, and converts the hydrogen stored in the hydrogen reservoir into electric power;
Equipped energy system.

Images